[Users] Ovirt Gluster problems
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --K1RorUOWmjDsov5Mv5auigJcOlmISEK1x Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi, I had some issues with a gluster cluster and after some time trying to get the storage domain up or delete it (I opened a BZ about a deadlock in the process of removing the domain) I gave up and destroyed the DC. The thing is that I want to add the hosts that where part of the DC and now I get that I can't as they have the volume. I try to stop the volume but I can't as no host is running in the deleted cluster and for some reason, ovirt needs that. I can't delete the hosts either as they have the volume... so I'm back in another chicken and egg problem. Any hints?? PD: I can't nuke the hole ovirt plataform as I have another DC in production.... otherwise I would :-) Regards, --K1RorUOWmjDsov5Mv5auigJcOlmISEK1x Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS5+vaAAoJEC5KDDg2+NMMHR4H/1kf5+MMK9VMbNn0cNLXOsDL pLcFHMAqq5+C5XvA0CrKgsEpTNvkjH49X3hB/ulhwDtXDJBLcoHVC+mAoqMmbkeQ 1bJaUCNQtst9BYOtJ4eq5ZXCGpB7rvEe5wZGuBeJqZUFnJ1FekVc3F1RXbfBmVeE ELQlJ0rp4cufZod8TwNh1O+wSx0nVHC9TJPKPSPA/qKMqrG2IIiR/3wGlFpeu3QJ eBcvixk7wf8YyxGvOwBDYVoc4+tYCHmqHiGAj6E6lDAVDySgvRTg/wH5OkoYGpn1 tSRiLWlSzd1A+XpKveOabPdBfOKqSoSUTfgq+NiwTD/9cKJuElb2g5F6UycjPAw= =oHX+ -----END PGP SIGNATURE----- --K1RorUOWmjDsov5Mv5auigJcOlmISEK1x--
Not sure if this is exactly your issue, but this post here: http://comments.gmane.org/gmane.comp.emulators.ovirt.user/12200 might lead you in the right direction. "one note - if you back it up while its attached to an engine, you will need to edit its meta data file to remove the association to allow the other engine to connect it to the new pool for restore." *Steve Dainard * On Tue, Jan 28, 2014 at 12:41 PM, Juan Pablo Lorier <jplorier@gmail.com>wrote:
Hi,
I had some issues with a gluster cluster and after some time trying to get the storage domain up or delete it (I opened a BZ about a deadlock in the process of removing the domain) I gave up and destroyed the DC. The thing is that I want to add the hosts that where part of the DC and now I get that I can't as they have the volume. I try to stop the volume but I can't as no host is running in the deleted cluster and for some reason, ovirt needs that. I can't delete the hosts either as they have the volume... so I'm back in another chicken and egg problem. Any hints??
PD: I can't nuke the hole ovirt plataform as I have another DC in production.... otherwise I would :-)
Regards,
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
This is a multi-part message in MIME format. --------------020406080501020400040905 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit On 01/29/2014 04:15 AM, Steve Dainard wrote:
Not sure if this is exactly your issue, but this post here: http://comments.gmane.org/gmane.comp.emulators.ovirt.user/12200 might lead you in the right direction.
"one note - if you back it up while its attached to an engine, you will need to edit its meta data file to remove the association to allow the other engine to connect it to the new pool for restore."
Did this solve your issue? If not, could you let us know the error messages from the logs? (engine.log) If you're looking to remove a host from a gluster cluster when there are no online hosts, checking the Force option should do this for you.
*Steve Dainard *
On Tue, Jan 28, 2014 at 12:41 PM, Juan Pablo Lorier <jplorier@gmail.com <mailto:jplorier@gmail.com>> wrote:
Hi,
I had some issues with a gluster cluster and after some time trying to get the storage domain up or delete it (I opened a BZ about a deadlock in the process of removing the domain) I gave up and destroyed the DC. The thing is that I want to add the hosts that where part of the DC and now I get that I can't as they have the volume. I try to stop the volume but I can't as no host is running in the deleted cluster and for some reason, ovirt needs that. I can't delete the hosts either as they have the volume... so I'm back in another chicken and egg problem. Any hints??
PD: I can't nuke the hole ovirt plataform as I have another DC in production.... otherwise I would :-)
Regards,
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
--------------020406080501020400040905 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit <html> <head> <meta content="text/html; charset=ISO-8859-1" http-equiv="Content-Type"> </head> <body bgcolor="#FFFFFF" text="#000000"> <br> <div class="moz-cite-prefix">On 01/29/2014 04:15 AM, Steve Dainard wrote:<br> </div> <blockquote cite="mid:CAHnsdUtE5TCQW77_yehJ1=Vh69JH_ZamtT0DtGizyGYyn-MRqQ@mail.gmail.com" type="cite"> <div dir="ltr">Not sure if this is exactly your issue, but this post here: <a moz-do-not-send="true" href="http://comments.gmane.org/gmane.comp.emulators.ovirt.user/12200">http://comments.gmane.org/gmane.comp.emulators.ovirt.user/12200</a> might lead you in the right direction. <div> <br> </div> <div> <pre style="line-height:24px;text-align:justify">"one note - if you back it up while its attached to an engine, you will need to edit its meta data file to remove the association to allow the other engine to connect it to the new pool for restore."</pre> </div> <div class="gmail_extra"><br clear="all"> </div> </div> </blockquote> <br> Did this solve your issue?<br> <br> If not, could you let us know the error messages from the logs? (engine.log)<br> If you're looking to remove a host from a gluster cluster when there are no online hosts, checking the Force option should do this for you.<br> <br> <blockquote cite="mid:CAHnsdUtE5TCQW77_yehJ1=Vh69JH_ZamtT0DtGizyGYyn-MRqQ@mail.gmail.com" type="cite"> <div dir="ltr"> <div class="gmail_extra"> <div> <div dir="ltr"><span style="font-family:arial,sans-serif;font-size:16px"><strong>Steve Dainard </strong></span><span style="font-size:12px"></span><br> <br> </div> </div> <br> <br> <div class="gmail_quote">On Tue, Jan 28, 2014 at 12:41 PM, Juan Pablo Lorier <span dir="ltr"><<a moz-do-not-send="true" href="mailto:jplorier@gmail.com" target="_blank" onclick="window.open('https://mail.google.com/mail/?view=cm&tf=1&to=jplorier@gmail.com&cc=&bcc=&su=&body=','_blank');return false;">jplorier@gmail.com</a>></span> wrote:<br> <blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br> <br> I had some issues with a gluster cluster and after some time trying to<br> get the storage domain up or delete it (I opened a BZ about a deadlock<br> in the process of removing the domain) I gave up and destroyed the DC.<br> The thing is that I want to add the hosts that where part of the DC and<br> now I get that I can't as they have the volume. I try to stop the volume<br> but I can't as no host is running in the deleted cluster and for some<br> reason, ovirt needs that.<br> I can't delete the hosts either as they have the volume... so I'm back<br> in another chicken and egg problem.<br> Any hints??<br> <br> PD: I can't nuke the hole ovirt plataform as I have another DC in<br> production.... otherwise I would :-)<br> <br> Regards,<br> <br> <br> _______________________________________________<br> Users mailing list<br> <a moz-do-not-send="true" href="mailto:Users@ovirt.org" onclick="window.open('https://mail.google.com/mail/?view=cm&tf=1&to=Users@ovirt.org&cc=&bcc=&su=&body=','_blank');return false;">Users@ovirt.org</a><br> <a moz-do-not-send="true" href="http://lists.ovirt.org/mailman/listinfo/users" target="_blank">http://lists.ovirt.org/mailman/listinfo/users</a><br> <br> </blockquote> </div> <br> </div> </div> <br> <fieldset class="mimeAttachmentHeader"></fieldset> <br> <pre wrap="">_______________________________________________ Users mailing list <a class="moz-txt-link-abbreviated" href="mailto:Users@ovirt.org">Users@ovirt.org</a> <a class="moz-txt-link-freetext" href="http://lists.ovirt.org/mailman/listinfo/users">http://lists.ovirt.org/mailman/listinfo/users</a> </pre> </blockquote> <br> </body> </html> --------------020406080501020400040905--
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --66emp7BA7bxAHcrH9o6vnBTPRuUUC9QbB Content-Type: multipart/alternative; boundary="------------040407060705020309060309" This is a multi-part message in MIME format. --------------040407060705020309060309 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi Sahina, I've tried to remove the volume also from gluster manually to see if ovirt would set it as inactive, but nothing. The thing now is that as I was unable to remove the gluster domain, I decided to force remove the hole data center and to my surprise, though it deleted the DC and all it's data, it didn't delete the volume and is still there in volume tab depending on a non existing cluster and thus I'm not able to do anything with it as everything fails with "there's no host in the cluster".... Sounds like the destroy operation is not checking everything and leaves things hanging in the air. Can you help me hacking the database to remove the volume? I'm in a rush and can't wait till this bug gets solved (didn't open a bz as I'm too busy right now). I have a deadline to get this working by the 10th February and won't be at the office from 4 through 6, so you may imagine where I'm standing. I think the hole logic of dependencies in the gluster DC (maybe more) must be checked as since I had my gluster domain down I was unable to do mostly anything in the DC because everything depended on each other. I couldn't delete the domain as there where vms in it, but couldn't delete the vms as they needed the domain. Couldn't remove the hosts as they where bricks, but couldn't remove from the volume as the domain was down, etc, etc Regards, On 29/01/14 07:28, Sahina Bose wrote:
On 01/29/2014 04:15 AM, Steve Dainard wrote:
Not sure if this is exactly your issue, but this post here: http://comments.gmane.org/gmane.comp.emulators.ovirt.user/12200 might lead you in the right direction.
"one note - if you back it up while its attached to an engine, you wil=
l=20
need to edit its meta data file to remove the association to allow the= =20 other engine to connect it to the new pool for restore."
Did this solve your issue?
If not, could you let us know the error messages from the logs? (engine.log) If you're looking to remove a host from a gluster cluster when there are no online hosts, checking the Force option should do this for you.
*Steve Dainard *
On Tue, Jan 28, 2014 at 12:41 PM, Juan Pablo Lorier <jplorier@gmail.com <mailto:jplorier@gmail.com>> wrote:
Hi,
I had some issues with a gluster cluster and after some time trying to get the storage domain up or delete it (I opened a BZ about a deadlock in the process of removing the domain) I gave up and destroyed the DC. The thing is that I want to add the hosts that where part of the DC and now I get that I can't as they have the volume. I try to stop the volume but I can't as no host is running in the deleted cluster and for s= ome reason, ovirt needs that. I can't delete the hosts either as they have the volume... so I'm back in another chicken and egg problem. Any hints??
PD: I can't nuke the hole ovirt plataform as I have another DC in production.... otherwise I would :-)
Regards,
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
--------------040407060705020309060309 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable <html> <head> <meta content=3D"text/html; charset=3DISO-8859-1" http-equiv=3D"Content-Type"> </head> <body text=3D"#000000" bgcolor=3D"#FFFFFF"> Hi Sahina,<br> <br> I've tried to remove the volume also from gluster manually to see if ovirt would set it as inactive, but nothing.<br> The thing now is that as I was unable to remove the gluster domain, I decided to force remove the hole data center and to my surprise, though it deleted the DC and all it's data, it didn't delete the volume and is still there in volume tab depending on a non existing cluster and thus I'm not able to do anything with it as everything fails with "there's no host in the cluster"....<br> Sounds like the destroy operation is not checking everything and leaves things hanging in the air.<br> Can you help me hacking the database to remove the volume? I'm in a rush and can't wait till this bug gets solved (didn't open a bz as I'm too busy right now).<br> I have a deadline to get this working by the 10th February and won't be at the office from 4 through 6, so you may imagine where I'm standing.<br> I think the hole logic of dependencies in the gluster DC (maybe more) must be checked as since I had my gluster domain down I was unable to do mostly anything in the DC because everything depended on each other. I couldn't delete the domain as there where vms in it, but couldn't delete the vms as they needed the domain. Couldn't remove the hosts as they where bricks, but couldn't remove from the volume as the domain was down, etc, etc<br> Regards,<br> <br> <div class=3D"moz-cite-prefix">On 29/01/14 07:28, Sahina Bose wrote:<= br> </div> <blockquote cite=3D"mid:52E8C9BB.9020103@redhat.com" type=3D"cite"> <meta content=3D"text/html; charset=3DISO-8859-1" http-equiv=3D"Content-Type"> <br> <div class=3D"moz-cite-prefix">On 01/29/2014 04:15 AM, Steve Dainar= d wrote:<br> </div> <blockquote cite=3D"mid:CAHnsdUtE5TCQW77_yehJ1=3DVh69JH_ZamtT0DtGizyGYyn-MRqQ@mail.gm= ail.com" type=3D"cite"> <div dir=3D"ltr">Not sure if this is exactly your issue, but this= post here: <a moz-do-not-send=3D"true" href=3D"http://comments.gmane.org/gmane.comp.emulators.ovirt.= user/12200">http://comments.gmane.org/gmane.comp.emulators.ovirt.user/122= 00</a> might lead you in the right direction. <div> <br> </div> <div> <pre style=3D"line-height:24px;text-align:justify">"one note = - if you back it up while its attached to an engine, you will=20 need to edit its meta data file to remove the association to allow the=20 other engine to connect it to the new pool for restore."</pre> </div> <div class=3D"gmail_extra"><br clear=3D"all"> </div> </div> </blockquote> <br> Did this solve your issue?<br> <br> If not, could you let us know the error messages from the logs? (engine.log)<br> If you're looking to remove a host from a gluster cluster when there are no online hosts, checking the Force option should do this for you.<br> <br> <blockquote cite=3D"mid:CAHnsdUtE5TCQW77_yehJ1=3DVh69JH_ZamtT0DtGizyGYyn-MRqQ@mail.gm= ail.com" type=3D"cite"> <div dir=3D"ltr"> <div class=3D"gmail_extra"> <div> <div dir=3D"ltr"><span style=3D"font-family:arial,sans-serif;font-size:16px"><= strong>Steve Dainard </strong></span><span style=3D"font-size:12px"></span><br> <br> </div> </div> <br> <br> <div class=3D"gmail_quote">On Tue, Jan 28, 2014 at 12:41 PM, Juan Pablo Lorier <span dir=3D"ltr"><<a moz-do-not-send=3D"true" href=3D"mailto:jplorier@gmail.com" target=3D"_blank" onclick=3D"window.open('https://mail.google.com/mail/?v= iew=3Dcm&tf=3D1&to=3Djplorier@gmail.com&cc=3D&bcc=3D&= su=3D&body=3D','_blank');return false;">jplorier@gmail.com</a>></span> wrote:<br> <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>= <br> I had some issues with a gluster cluster and after some time trying to<br> get the storage domain up or delete it (I opened a BZ about a deadlock<br> in the process of removing the domain) I gave up and destroyed the DC.<br> The thing is that I want to add the hosts that where part of the DC and<br> now I get that I can't as they have the volume. I try to stop the volume<br> but I can't as no host is running in the deleted cluster and for some<br> reason, ovirt needs that.<br> I can't delete the hosts either as they have the volume... so I'm back<br> in another chicken and egg problem.<br> Any hints??<br> <br> PD: I can't nuke the hole ovirt plataform as I have another DC in<br> production.... otherwise I would :-)<br> <br> Regards,<br> <br> <br> _______________________________________________<br> Users mailing list<br> <a moz-do-not-send=3D"true" href=3D"mailto:Users@ovirt.or= g" onclick=3D"window.open('https://mail.google.com/mail/?v= iew=3Dcm&tf=3D1&to=3DUsers@ovirt.org&cc=3D&bcc=3D&su=3D= &body=3D','_blank');return false;">Users@ovirt.org</a><br> <a moz-do-not-send=3D"true" href=3D"http://lists.ovirt.org/mailman/listinfo/users" target=3D"_blank">http://lists.ovirt.org/mailman/listin= fo/users</a><br> <br> </blockquote> </div> <br> </div> </div> <br> <fieldset class=3D"mimeAttachmentHeader"></fieldset> <br> <pre wrap=3D"">_______________________________________________ Users mailing list <a moz-do-not-send=3D"true" class=3D"moz-txt-link-abbreviated" href=3D"ma= ilto:Users@ovirt.org">Users@ovirt.org</a> <a moz-do-not-send=3D"true" class=3D"moz-txt-link-freetext" href=3D"http:= //lists.ovirt.org/mailman/listinfo/users">http://lists.ovirt.org/mailman/= listinfo/users</a> </pre> </blockquote> <br> </blockquote> <br> </body> </html> --------------040407060705020309060309-- --66emp7BA7bxAHcrH9o6vnBTPRuUUC9QbB Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS6qZRAAoJEC5KDDg2+NMM/awH/jiNwA0MBA6+MahX/puviZbf rdFLPO8252isL0SgOqe3UDyXE6QuK/idR0/Y/n0QFdERauwB5uEs3CwQbRUrPwC8 xKbH9c/kfcNMGiJUHwouNjGIYc2c//Am9JEhSG1MpM2MEJtEUiQedmKVAwLxansH wV2q7UBys7xCJNMZN6VEODMaH3nQzdPS5FLNyaa1BNU/SRIhp6ly0Kw/rGeNOzmV tYoNEJ8jC1vYomdU2TOKKJf9naeq5BmM0PW/5tcu5aBpjz1NzpGYXnvpw2U0CZsr fsbedw/0Z7jm93BuWvwy+htNw1CRXD4y0wFUFGGny2hyBsjEUl9JwqDDrwXgOzQ= =lf40 -----END PGP SIGNATURE----- --66emp7BA7bxAHcrH9o6vnBTPRuUUC9QbB--
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --o8g9s6BodJt91C58VR4p2bK4UQrTsW1Ot Content-Type: multipart/alternative; boundary="------------080804060407070207070609" This is a multi-part message in MIME format. --------------080804060407070207070609 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi Sahina, it's been some days since I got this to work thanks to your help. I still don't have the time to seek trhough the logs to find when the engine left things without erasing after the force deletion of the DC, but if you wish, I can send you the raw logs and you may search in them. Tell me if it's ok with you and I'll send you a google drive link. Regards, On 29/01/14 07:28, Sahina Bose wrote:
On 01/29/2014 04:15 AM, Steve Dainard wrote:
Not sure if this is exactly your issue, but this post here: http://comments.gmane.org/gmane.comp.emulators.ovirt.user/12200 might lead you in the right direction.
"one note - if you back it up while its attached to an engine, you wil=
l=20
need to edit its meta data file to remove the association to allow the= =20 other engine to connect it to the new pool for restore."
Did this solve your issue?
If not, could you let us know the error messages from the logs? (engine.log) If you're looking to remove a host from a gluster cluster when there are no online hosts, checking the Force option should do this for you.
*Steve Dainard *
On Tue, Jan 28, 2014 at 12:41 PM, Juan Pablo Lorier <jplorier@gmail.com <mailto:jplorier@gmail.com>> wrote:
Hi,
I had some issues with a gluster cluster and after some time trying to get the storage domain up or delete it (I opened a BZ about a deadlock in the process of removing the domain) I gave up and destroyed the DC. The thing is that I want to add the hosts that where part of the DC and now I get that I can't as they have the volume. I try to stop the volume but I can't as no host is running in the deleted cluster and for s= ome reason, ovirt needs that. I can't delete the hosts either as they have the volume... so I'm back in another chicken and egg problem. Any hints??
PD: I can't nuke the hole ovirt plataform as I have another DC in production.... otherwise I would :-)
Regards,
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
--------------080804060407070207070609 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable <html> <head> <meta content=3D"text/html; charset=3DISO-8859-1" http-equiv=3D"Content-Type"> </head> <body text=3D"#000000" bgcolor=3D"#FFFFFF"> Hi Sahina,<br> <br> it's been some days since I got this to work thanks to your help. I still don't have the time to seek trhough the logs to find when the engine left things without erasing after the force deletion of the DC, but if you wish, I can send you the raw logs and you may search in them.<br> Tell me if it's ok with you and I'll send you a google drive link.<br=
Regards,<br> <br> <div class=3D"moz-cite-prefix">On 29/01/14 07:28, Sahina Bose wrote:<= br> </div> <blockquote cite=3D"mid:52E8C9BB.9020103@redhat.com" type=3D"cite"> <meta content=3D"text/html; charset=3DISO-8859-1" http-equiv=3D"Content-Type"> <br> <div class=3D"moz-cite-prefix">On 01/29/2014 04:15 AM, Steve Dainar= d wrote:<br> </div> <blockquote cite=3D"mid:CAHnsdUtE5TCQW77_yehJ1=3DVh69JH_ZamtT0DtGizyGYyn-MRqQ@mail.gm= ail.com" type=3D"cite"> <div dir=3D"ltr">Not sure if this is exactly your issue, but this= post here: <a moz-do-not-send=3D"true" href=3D"http://comments.gmane.org/gmane.comp.emulators.ovirt.= user/12200">http://comments.gmane.org/gmane.comp.emulators.ovirt.user/122= 00</a> might lead you in the right direction. <div> <br> </div> <div> <pre style=3D"line-height:24px;text-align:justify">"one note = - if you back it up while its attached to an engine, you will=20 need to edit its meta data file to remove the association to allow the=20 other engine to connect it to the new pool for restore."</pre> </div> <div class=3D"gmail_extra"><br clear=3D"all"> </div> </div> </blockquote> <br> Did this solve your issue?<br> <br> If not, could you let us know the error messages from the logs? (engine.log)<br> If you're looking to remove a host from a gluster cluster when there are no online hosts, checking the Force option should do this for you.<br> <br> <blockquote cite=3D"mid:CAHnsdUtE5TCQW77_yehJ1=3DVh69JH_ZamtT0DtGizyGYyn-MRqQ@mail.gm= ail.com" type=3D"cite"> <div dir=3D"ltr"> <div class=3D"gmail_extra"> <div> <div dir=3D"ltr"><span style=3D"font-family:arial,sans-serif;font-size:16px"><= strong>Steve Dainard </strong></span><span style=3D"font-size:12px"></span><br> <br> </div> </div> <br> <br> <div class=3D"gmail_quote">On Tue, Jan 28, 2014 at 12:41 PM, Juan Pablo Lorier <span dir=3D"ltr"><<a moz-do-not-send=3D"true" href=3D"mailto:jplorier@gmail.com" target=3D"_blank" onclick=3D"window.open('https://mail.google.com/mail/?v= iew=3Dcm&tf=3D1&to=3Djplorier@gmail.com&cc=3D&bcc=3D&= su=3D&body=3D','_blank');return false;">jplorier@gmail.com</a>></span> wrote:<br> <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br>= <br> I had some issues with a gluster cluster and after some time trying to<br> get the storage domain up or delete it (I opened a BZ about a deadlock<br> in the process of removing the domain) I gave up and destroyed the DC.<br> The thing is that I want to add the hosts that where part of the DC and<br> now I get that I can't as they have the volume. I try to stop the volume<br> but I can't as no host is running in the deleted cluster and for some<br> reason, ovirt needs that.<br> I can't delete the hosts either as they have the volume... so I'm back<br> in another chicken and egg problem.<br> Any hints??<br> <br> PD: I can't nuke the hole ovirt plataform as I have another DC in<br> production.... otherwise I would :-)<br> <br> Regards,<br> <br> <br> _______________________________________________<br> Users mailing list<br> <a moz-do-not-send=3D"true" href=3D"mailto:Users@ovirt.or= g" onclick=3D"window.open('https://mail.google.com/mail/?v= iew=3Dcm&tf=3D1&to=3DUsers@ovirt.org&cc=3D&bcc=3D&su=3D= &body=3D','_blank');return false;">Users@ovirt.org</a><br> <a moz-do-not-send=3D"true" href=3D"http://lists.ovirt.org/mailman/listinfo/users" target=3D"_blank">http://lists.ovirt.org/mailman/listin= fo/users</a><br> <br> </blockquote> </div> <br> </div> </div> <br> <fieldset class=3D"mimeAttachmentHeader"></fieldset> <br> <pre wrap=3D"">_______________________________________________ Users mailing list <a moz-do-not-send=3D"true" class=3D"moz-txt-link-abbreviated" href=3D"ma= ilto:Users@ovirt.org">Users@ovirt.org</a> <a moz-do-not-send=3D"true" class=3D"moz-txt-link-freetext" href=3D"http:= //lists.ovirt.org/mailman/listinfo/users">http://lists.ovirt.org/mailman/= listinfo/users</a> </pre> </blockquote> <br> </blockquote> <br> </body> </html> --------------080804060407070207070609-- --o8g9s6BodJt91C58VR4p2bK4UQrTsW1Ot Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS+RMJAAoJEC5KDDg2+NMMv+IH/ReQDO5RoK4eZm/Dx/T594qZ 6kXDhpBkyedgFGfz4Pfv7DYd9fkGXN3avLF7PYdfPH6dmYAOFm8jtgbCena08otF BpYj/qnjbP1bMUdDhb3aP9lnfO9dmep/U2SMXHfIUqxccJzR1UTH1e19Ta6wqg4a aTjO8m0EUA2hOSrZcgF0bo9mObLx46Y50s53zRUIKDefnQhwl1roGpi96CLEVhwC nzXszYQfoyWSdWvUStrIfnyyMQrW1vUHB8I3i43/u2GiMUcNKEkHXadCdGppTOdh P21mLlIL4/BI1KXo00RzJEMB0uF1RVs1HsV81lvGrT5/7j5W2p0Hi5tiKeQVs4g= =oFxn -----END PGP SIGNATURE----- --o8g9s6BodJt91C58VR4p2bK4UQrTsW1Ot--
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --jJk9qpC3CG6D5ARwaQd9JL3g4r3t35RbJ Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi Sahina, Sorry to ping you directly, but you were of great help last time. I'm also copying the list. In the process of recreating the DC I've decided to migrate the engine from a virtualbox vm to a bare metal host and I can't get engine-backup to restore the engine. Both are Centos 6.5 with engine 3.3.3, but in the new box I can't log to the database with psql -U engine -W The error is: [root@centos-ovirt ~]# engine-backup --mode=3Drestore --file=3D/root/backup_20140212.bkp --log=3D/root/restore.log Restoring... FATAL: Can't connect to the database I've added to pg_hba.conf the lines I found in the web: host all all md5 and tryied with the conf from the one that works but nothing. Any clues? Regards, --jJk9qpC3CG6D5ARwaQd9JL3g4r3t35RbJ Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS+6jgAAoJEC5KDDg2+NMM5JsIANDR5o94/OaN+ax1Tg6iQ9xm S6mRQTbxPHXISgEj1QniWegV+ovv9jm/2HzOCmkzjIHX0aBWlpat3Iu2rzEiH3CQ XQVeBDgRAoKise2fPki+R6wAaAYqRf+Y54f7/jf/aZyobF01KndKfOTREVg4naMo QbL4HURSgf9iaMIC4xY7YbZIyLoxxFnM0u4hO8MTPLKPe8QF9zzNzt32ZNSoBod4 PIlpw1rhIkkBjkmtmR7t83FphlOSdxadFwftc7GTqobB41vVMscFuo+X8IyaFFi6 Rs7jdRtnrtvR1zq1ugbbSH24XeScHW73GXuNhp2dU/ZLdCXf5P6KimxGBSQKO+Y= =k5BY -----END PGP SIGNATURE----- --jJk9qpC3CG6D5ARwaQd9JL3g4r3t35RbJ--
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Sahina Bose" <sabose@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:01:20 PM Subject: [Users] Problems accesing the database
Hi Sahina,
Sorry to ping you directly, but you were of great help last time. I'm also copying the list. In the process of recreating the DC I've decided to migrate the engine from a virtualbox vm to a bare metal host and I can't get engine-backup to restore the engine. Both are Centos 6.5 with engine 3.3.3, but in the new box I can't log to the database with psql -U engine -W The error is:
[root@centos-ovirt ~]# engine-backup --mode=restore --file=/root/backup_20140212.bkp --log=/root/restore.log Restoring... FATAL: Can't connect to the database
I've added to pg_hba.conf the lines I found in the web:
host all all md5
and tryied with the conf from the one that works but nothing. Any clues?
Unlike engine-setup, engine-backup does nothing for you regarding creating the database - it's basically the same as creating a database to be used for a remote-db setup. See e.g. [1] for details, also 'engine-backup --help'. [1] http://www.ovirt.org/OVirt_Engine_Development_Environment#Database -- Didi
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --mIHNOnbAdv7XaOQQMDkivlTdk39pnJU5G Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi Yedidyah, But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate Regards, On 12/02/14 15:44, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Sahina Bose" <sabose@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:01:20 PM Subject: [Users] Problems accesing the database
Hi Sahina,
Sorry to ping you directly, but you were of great help last time. I'm also copying the list. In the process of recreating the DC I've decided to migrate the engine=
from a virtualbox vm to a bare metal host and I can't get engine-backu= p to restore the engine. Both are Centos 6.5 with engine 3.3.3, but in the new box I can't log = to the database with psql -U engine -W The error is:
[root@centos-ovirt ~]# engine-backup --mode=3Drestore --file=3D/root/backup_20140212.bkp --log=3D/root/restore.log Restoring... FATAL: Can't connect to the database
I've added to pg_hba.conf the lines I found in the web:
host all all md5
and tryied with the conf from the one that works but nothing. Any clues? Unlike engine-setup, engine-backup does nothing for you regarding creating the database - it's basically the same as creating a database to be used for a remote-db setup. See e.g. [1] for details, also 'engine-backup --help'.
[1] http://www.ovirt.org/OVirt_Engine_Development_Environment#Database
--mIHNOnbAdv7XaOQQMDkivlTdk39pnJU5G Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS+7WXAAoJEC5KDDg2+NMMEE0IAKF2jvNEtrs0kf7BllGcT52N UJ8YAGrCdWMUZ+A4gf/LAy7ehDtlPCuLDhrVgmUDVBggruU0EcVPKIEI5gtjtK2t MWw/3ADb4iYXg467ZXZ8GlAYhzz1i/WSwJyJYGRQIaHlcRG7xvFhekCsSOJBrhlK 0Z9IeG+kkFjhQWzm+V5WUkeW5Uh2E44aWd7gNfMStmZIaw9/AATelRWhwxoyelyu Js+QweuVcQYyL0flmwIB2tfHBAcuFWrOiVXkMib9l/uCeGqquzIOB0pEhRPnkaiI 5YMWi7MMD0B8NpCHP+1N0Z2CQfW3nCwJokzrUbGseIOpgMoxq9C69MdMbQCl4KM= =SCAu -----END PGP SIGNATURE----- --mIHNOnbAdv7XaOQQMDkivlTdk39pnJU5G--
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate
No. There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=backup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=restore Why does this work? Because 'engine-cleanup', since 3.3, does not drop the database nor user inside postgres. So when restore tries to access this database using this user and password it succeeds. In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool, you end up almost ready, but not quite - because the password is random, and therefore different between the installations. In principle you could have provided just the password to restore, but we decided that if you need to change the credentials, you should pass all of them (except for defaults). Hope this clarifies, -- Didi
I tried this after upgrading to 3.4.0-beta2 and then attempting to downgrade back to 3.3.3. I performed an engine-cleanup, accepting defaults. My backup on 3.3.3 was $ engine-backup --mode=backup --scope=all --file=engine-20140211-1457.tar.bz2 --log=engine-backup.log $ engine-cleanup <log attached> $ engine-backup --mode=restore --scope=all --file=engine-20140211-1457.tar.bz2 --log=engine-backup.log Restoring... FATAL: Database is not empty What next? - Trey On Wed, Feb 12, 2014 at 12:03 PM, Yedidyah Bar David <didi@redhat.com> wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate
No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=backup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=restore
Why does this work? Because 'engine-cleanup', since 3.3, does not drop the database nor user inside postgres. So when restore tries to access this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool, you end up almost ready, but not quite - because the password is random, and therefore different between the installations. In principle you could have provided just the password to restore, but we decided that if you need to change the credentials, you should pass all of them (except for defaults).
Hope this clarifies, -- Didi _______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --v5c3oNXSCPUgD3stLT4bbIJo68u5GsQkN Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Almost, when you mean random password, you mean for the engine user in the database? The new install and the old differ in some things in the db, so I tried to made then the same, but I saw a post of a user that removed the engine database and user and created them again before trying to restore, is that a good path? Regards, On 12/02/14 16:03, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=3Dbackup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=3Drestore
Why does this work? Because 'engine-cleanup', since 3.3, does not drop the database nor user inside postgres. So when restore tries to access this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool, you end up almost ready, but not quite - because the password is random= , and therefore different between the installations. In principle you cou= ld have provided just the password to restore, but we decided that if you need to change the credentials, you should pass all of them (except for=
defaults).
Hope this clarifies,
--v5c3oNXSCPUgD3stLT4bbIJo68u5GsQkN Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS+7yeAAoJEC5KDDg2+NMMPEEIALsW3vjK+IBcCAWoX+EjahHu 0FQpa5Lqk2vlz74LH9gU4Q42jU12sYwi1mz9vb6zgtUj1tR+zqK56iwjIqjyOSY4 hyLbKcQCe9QJNbBT+m48lYzhZsdvJypaEIEIaHoAt4lT9rQUmMvVCCKSC6NadNbS Uy1MCU9Qg6jCcUFG4EXguOTfZaAfKkHbl/Vm58C+HidZT5W0I41AJVav0QzWG5GJ 6Fu25rxtWGMXQsqofre8bCPhp3eaHp2A06ldMJpR4lV7VV19o9+tu3c1BTC3+w9i hSWSu9XJnYMZjl1zmJIPYba1+z45ovKjAlgH/JvKBu+kJ1EwWiPY25k7UeIaHNY= =4e9Z -----END PGP SIGNATURE----- --v5c3oNXSCPUgD3stLT4bbIJo68u5GsQkN--
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --a8BfA7x13RBwSaKFklSkrhkEWxmcqvQAi Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable OK, that did work. I droped the database and created it back and also changed the password for the engine user. That did the trick. I now have to figure out why the web is blank, but at least is a step forward. Regards, On 12/02/14 16:03, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=3Dbackup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=3Drestore
Why does this work? Because 'engine-cleanup', since 3.3, does not drop the database nor user inside postgres. So when restore tries to access this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool, you end up almost ready, but not quite - because the password is random= , and therefore different between the installations. In principle you cou= ld have provided just the password to restore, but we decided that if you need to change the credentials, you should pass all of them (except for=
defaults).
Hope this clarifies,
--a8BfA7x13RBwSaKFklSkrhkEWxmcqvQAi Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS+8PnAAoJEC5KDDg2+NMM2C0IAJfxvXWP9b/H+fNWdpKUoDF4 g8LmobdqOsd3uJSmaYyX0wLDdq3n1+vDfX8Z7ho5IrCz6+H4ZxqttvuH2fAaOh+x KjROfMRrTKH/G+P725AXWAnEI4tlCfXhRB938iKh+cAsAGCqe5eLwgsY0YsbVIGI dXp64GKya7kMQwCiq+ligptnMDde1oe8hR5t9iMgKr8Kh3oFSMVE0eMWzmqAmMKH FZbW7PLzMkjbnuc/Yi8VFv/9iALyT2Th5OAn9/tS2mndQt2CMCB9oKtkTGrRxh7+ i0G7WNLQlW7GeHWrEhlg9aYUTNtiurmz71MUsaQBsuKlVpStUCKulk3ynXSB5YM= =diMT -----END PGP SIGNATURE----- --a8BfA7x13RBwSaKFklSkrhkEWxmcqvQAi--
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --qvWbeLRdRGc72B6K9Cntojv7DP26SgCto Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Well, too soon to say boodbye. Thought I used --change-db-credentials in the restore, the engine seems to be unable to connect to the database. I assume that it didn't get the new password, so, is there a way to tell the engine about the new passwor= d? Regards, On 12/02/14 16:03, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=3Dbackup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=3Drestore
Why does this work? Because 'engine-cleanup', since 3.3, does not drop the database nor user inside postgres. So when restore tries to access this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool, you end up almost ready, but not quite - because the password is random= , and therefore different between the installations. In principle you cou= ld have provided just the password to restore, but we decided that if you need to change the credentials, you should pass all of them (except for=
defaults).
Hope this clarifies,
--qvWbeLRdRGc72B6K9Cntojv7DP26SgCto Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS+8czAAoJEC5KDDg2+NMMTD4H/AuPXcDPxDfh9lka535o7d+Q ncI7YkPYU+mdn5UylRZ0rJwNa8ixKhByy69HcDsE1AxMUwHTTmUrxNOsAwtcE0Gk 6l2lu57F9ZD1etnB1ni6GPLLmexV7GAvBCeut2AXfCudDEqRh37qfR2cBZQgX1uN XtA+bQA47+s9w4Ej3Wj8K88FTbFtaMi5fa16ZsUhuEC9IFOaHT0bCL7YQWLMStHP VnPhuTaxjQXSo4ZaqyILalzEChv5y8qm8eDm8huue+4MYz9OLSWhkRAe9kbYaYcz UGBk/kKVwG5SrvPWkZNCPC/g6LaliYBRqPvQGiAu8LW4w6QAIfFItyxZoKWQfeI= =9qzR -----END PGP SIGNATURE----- --qvWbeLRdRGc72B6K9Cntojv7DP26SgCto--
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 9:10:43 PM Subject: Re: [Users] engine-backup --restore
Well, too soon to say boodbye. Thought I used --change-db-credentials in the restore, the engine seems to be unable to connect to the database. I assume that it didn't get the new password, so, is there a way to tell the engine about the new password? Regards,
/etc/ovirt-engine/engine.conf/10-setup-database.conf or similar. Look for ENGINE_DB_* variables.
On 12/02/14 16:03, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=backup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=restore
Why does this work? Because 'engine-cleanup', since 3.3, does not drop the database nor user inside postgres. So when restore tries to access this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool, you end up almost ready, but not quite - because the password is random, and therefore different between the installations. In principle you could have provided just the password to restore, but we decided that if you need to change the credentials, you should pass all of them (except for defaults).
Hope this clarifies,
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
----- Original Message -----
From: "Alon Bar-Lev" <alonbl@redhat.com> To: "Juan Pablo Lorier" <jplorier@gmail.com> Cc: "Yedidyah Bar David" <didi@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 9:13:30 PM Subject: Re: [Users] engine-backup --restore
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 9:10:43 PM Subject: Re: [Users] engine-backup --restore
Well, too soon to say boodbye. Thought I used --change-db-credentials in the restore, the engine seems to be unable to connect to the database. I assume that it didn't get the new password, so, is there a way to tell the engine about the new password? Regards,
/etc/ovirt-engine/engine.conf/10-setup-database.conf or similar. Look for ENGINE_DB_* variables.
Indeed, but this should happen automatically if you passed correct credentials and restore succeeded. See e.g. the other report posted. Did restore succeed? Thanks, -- Didi
----- Original Message -----
From: "Alon Bar-Lev" <alonbl@redhat.com> To: "Juan Pablo Lorier" <jplorier@gmail.com> Cc: "Yedidyah Bar David" <didi@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 9:13:30 PM Subject: Re: [Users] engine-backup --restore
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 9:10:43 PM Subject: Re: [Users] engine-backup --restore
Well, too soon to say boodbye. Thought I used --change-db-credentials in the restore, the engine see= ms to be unable to connect to the database. I assume that it didn't get =
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --Hw0sn29FgA9AcbfJ2OrSOj1qK9fQhxcTl Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Hi David, It ended correctly and password was changed in the 10-setup-database.conf but there were still errors in the engine log pointing to an authentication problem against the database. I'm starting over again with some hints from another post, lets see what happens. Regards, On 13/02/14 04:23, Yedidyah Bar David wrote: the
new password, so, is there a way to tell the engine about the new pas= sword? Regards, /etc/ovirt-engine/engine.conf/10-setup-database.conf or similar. Look for ENGINE_DB_* variables. Indeed, but this should happen automatically if you passed correct credentials and restore succeeded. See e.g. the other report posted. Did restore succeed?
Thanks,
--Hw0sn29FgA9AcbfJ2OrSOj1qK9fQhxcTl Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS/OKeAAoJEC5KDDg2+NMMXXsIANATSbPvmcGAZ3+6B22hDztr 4un/09fSTUNPPwKNvipTJrc2lKDbvpwdYSdlHtT62ar7cg+QatElTwYJBMEVx1cX 0XGQ15ukuzdEW4X7OC5gxjO9DTEtROqt0PcAEPp89s27YHcSP54vhZD7KGY82T6y UTkN+iFL0XJG2/Oqh/nkFLj/sos9Y+fyR5aXswukNDQ7YMUlPZyf36367hNzCZq2 /TZpWTLe/xAKzv693yZu1H2hqe65zpNqGcNjZZ0ouWM8wwDp8pH6RUT9Dx6vODYe OWwP8Wq2iyK0LbwnX2tKU8X715+trjDyeEmpvcrqTNG4thifeBT5lbyCN5czIIY= =MkEe -----END PGP SIGNATURE----- --Hw0sn29FgA9AcbfJ2OrSOj1qK9fQhxcTl--
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --euTEL0bDdeo0xTvs8uoh21Tuv7u73bRqD Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Thanks Alon, you were right. On 12/02/14 17:13, Alon Bar-Lev wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 9:10:43 PM Subject: Re: [Users] engine-backup --restore
Well, too soon to say boodbye. Thought I used --change-db-credentials in the restore, the engine seem=
s
to be unable to connect to the database. I assume that it didn't get t= he new password, so, is there a way to tell the engine about the new pass= word? Regards, /etc/ovirt-engine/engine.conf/10-setup-database.conf or similar. Look for ENGINE_DB_* variables.
On 12/02/14 16:03, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it=
import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=3Dbackup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=3Drestore
Why does this work? Because 'engine-cleanup', since 3.3, does not dro= p the database nor user inside postgres. So when restore tries to acces= s this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool,=
you end up almost ready, but not quite - because the password is rand= om, and therefore different between the installations. In principle you c= ould have provided just the password to restore, but we decided that if yo= u need to change the credentials, you should pass all of them (except f= or defaults).
Hope this clarifies,
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
--euTEL0bDdeo0xTvs8uoh21Tuv7u73bRqD Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS/OHGAAoJEC5KDDg2+NMMq9MIAMZjoMybfwupPnHQXb2/0aK4 ram0lufxCiRlfoDrG/CeVfEtOv72g0PZW0Gf07eXSq9YGbaLBdI0FWLwh6pE5bu/ I/FHS4PUG80otHCygFZJMEiW2rnNh1ByoMnWHpwIl8fW7qgxPvQ+/DqpE6Wo3X+9 1qfBO5rOp+nxHcsE1Sv3UDDW8M1nX/urr5utNKhqkbBFrDoXkH8+crY5i4Fy63U6 2Kf3GV7qWZCf/Vv4FyrGlloe3ZXeoP6jNcJDpqjbiyu1wHRha97LZf6MfdI1CRyb 23pUrWa5YJrlHh64ng45TVX/gaFxecbIZ39ek9BA7x8MhXe0oUUaEHfCWTVnaxc= =JtRm -----END PGP SIGNATURE----- --euTEL0bDdeo0xTvs8uoh21Tuv7u73bRqD--
I was having the same issue (and posted about it today, with full steps). I'd reply to my current post to list but I don't seem to receive my own posts. I have since taken these steps (as root) $ su - postgres -c "dropdb engine" $ su - postgres -c "psql -c \"create user engine password '<PASSWORD>'\"" $ su - postgres -c "psql -c \"create database engine owner engine template template0 encoding 'UTF8' lc_collate 'en_US.UTF-8' lc_ctype 'en_US.UTF-8'\"" $ engine-backup --mode=restore --scope=all --file=engine-20140211-1457.tar.bz2 --log=engine-backup.log --change-db-credentials --db-host=localhost --db-port=5432 --db-user=engine --db-name=engine --db-password=<PASSWORD> Restoring... Rewriting /etc/ovirt-engine/engine.conf.d/10-setup-database.conf Note: you might need to manually fix: - iptables/firewalld configuration - autostart of ovirt-engine service You can now start the engine service and then restart httpd Done. $ engine-setup I believe I initially created the database incorrectly (ran 'createdb engine' as postgres user). I was getting errors during engine-setup that indicated the database could not be accessed. After the steps above, everything looks good. - Trey On Wed, Feb 12, 2014 at 1:10 PM, Juan Pablo Lorier <jplorier@gmail.com> wrote:
Well, too soon to say boodbye. Thought I used --change-db-credentials in the restore, the engine seems to be unable to connect to the database. I assume that it didn't get the new password, so, is there a way to tell the engine about the new password? Regards,
On 12/02/14 16:03, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=backup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=restore
Why does this work? Because 'engine-cleanup', since 3.3, does not drop the database nor user inside postgres. So when restore tries to access this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool, you end up almost ready, but not quite - because the password is random, and therefore different between the installations. In principle you could have provided just the password to restore, but we decided that if you need to change the credentials, you should pass all of them (except for defaults).
Hope this clarifies,
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
----- Original Message -----
From: "Trey Dockendorf" <treydock@gmail.com> To: "Juan Pablo Lorier" <jplorier@gmail.com> Cc: "Yedidyah Bar David" <didi@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 9:32:55 PM Subject: Re: [Users] engine-backup --restore
I was having the same issue (and posted about it today, with full steps). I'd reply to my current post to list but I don't seem to receive my own posts.
I have since taken these steps (as root)
$ su - postgres -c "dropdb engine" $ su - postgres -c "psql -c \"create user engine password '<PASSWORD>'\"" $ su - postgres -c "psql -c \"create database engine owner engine template template0 encoding 'UTF8' lc_collate 'en_US.UTF-8' lc_ctype 'en_US.UTF-8'\"" $ engine-backup --mode=restore --scope=all --file=engine-20140211-1457.tar.bz2 --log=engine-backup.log --change-db-credentials --db-host=localhost --db-port=5432 --db-user=engine --db-name=engine --db-password=<PASSWORD> Restoring... Rewriting /etc/ovirt-engine/engine.conf.d/10-setup-database.conf Note: you might need to manually fix: - iptables/firewalld configuration - autostart of ovirt-engine service You can now start the engine service and then restart httpd Done.
$ engine-setup
I believe I initially created the database incorrectly (ran 'createdb engine' as postgres user). I was getting errors during engine-setup that indicated the database could not be accessed.
After the steps above, everything looks good.
Good job. Thanks for the report! -- Didi
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --mEvAgst0dIjlkjhP3giOcIRW6QCC31tgL Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi Trey, Following your procedure, I was able to get the engine running. THANKS A LOT TO EVERYONE!! If you allow me, I'll create a wiki page with this mentioning you so others can get this easily. Regards, On 12/02/14 17:32, Trey Dockendorf wrote:
I was having the same issue (and posted about it today, with full steps). I'd reply to my current post to list but I don't seem to receive my own posts.
I have since taken these steps (as root)
$ su - postgres -c "dropdb engine" $ su - postgres -c "psql -c \"create user engine password '<PASSWORD>'\= "" $ su - postgres -c "psql -c \"create database engine owner engine template template0 encoding 'UTF8' lc_collate 'en_US.UTF-8' lc_ctype 'en_US.UTF-8'\"" $ engine-backup --mode=3Drestore --scope=3Dall --file=3Dengine-20140211-1457.tar.bz2 --log=3Dengine-backup.log --change-db-credentials --db-host=3Dlocalhost --db-port=3D5432 --db-user=3Dengine --db-name=3Dengine --db-password=3D<PASSWORD> Restoring... Rewriting /etc/ovirt-engine/engine.conf.d/10-setup-database.conf Note: you might need to manually fix: - iptables/firewalld configuration - autostart of ovirt-engine service You can now start the engine service and then restart httpd Done.
$ engine-setup
I believe I initially created the database incorrectly (ran 'createdb engine' as postgres user). I was getting errors during engine-setup that indicated the database could not be accessed.
After the steps above, everything looks good.
- Trey
On Wed, Feb 12, 2014 at 1:10 PM, Juan Pablo Lorier <jplorier@gmail.com>= wrote:
Well, too soon to say boodbye. Thought I used --change-db-credentials in the restore, the engine seem= s to be unable to connect to the database. I assume that it didn't get t= he new password, so, is there a way to tell the engine about the new pass= word? Regards,
On 12/02/14 16:03, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it=
import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=3Dbackup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=3Drestore
Why does this work? Because 'engine-cleanup', since 3.3, does not dro= p the database nor user inside postgres. So when restore tries to acces= s this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool,=
you end up almost ready, but not quite - because the password is rand= om, and therefore different between the installations. In principle you c= ould have provided just the password to restore, but we decided that if yo= u need to change the credentials, you should pass all of them (except f= or defaults).
Hope this clarifies,
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
--mEvAgst0dIjlkjhP3giOcIRW6QCC31tgL Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS/ObHAAoJEC5KDDg2+NMMJHoIALPzqv+wOBSBYvPTVVOYyDwq AmlNsLovYAxxtLWb1Vek6UtJPU3uSW3A48o33NhG43rBrrbayKYYMWiC/+mPUr18 i3UFzbETYK+iN79iBvBoVkpgxt7k1TxvHmP1Qr34sBweT+8z3ZsJyIK7GykMgDEU qV6wFHnJmdMLqo3RUdWPSBAQWlLBrnogBa/ATd0X/LTdImU9BFsF0S05Z1+zXQ/H tccyTtOPDnhmEyCsTsxTPuH+IAKo4Ripiogh9RRot3KBuPziLnquE/eUbfIzNMX1 lj9H17e8vP1/g4mOM8+/x7gR+TLmqsMHL/Jjc8w2JwLuz1FX+11MbiNGFg5ydKA= =fjv6 -----END PGP SIGNATURE----- --mEvAgst0dIjlkjhP3giOcIRW6QCC31tgL--
A wiki page is a good idea, glad the steps worked in more than my case. - Trey On Thu, Feb 13, 2014 at 9:37 AM, Juan Pablo Lorier <jplorier@gmail.com> wrote:
Hi Trey,
Following your procedure, I was able to get the engine running. THANKS A LOT TO EVERYONE!! If you allow me, I'll create a wiki page with this mentioning you so others can get this easily. Regards,
On 12/02/14 17:32, Trey Dockendorf wrote:
I was having the same issue (and posted about it today, with full steps). I'd reply to my current post to list but I don't seem to receive my own posts.
I have since taken these steps (as root)
$ su - postgres -c "dropdb engine" $ su - postgres -c "psql -c \"create user engine password '<PASSWORD>'\"" $ su - postgres -c "psql -c \"create database engine owner engine template template0 encoding 'UTF8' lc_collate 'en_US.UTF-8' lc_ctype 'en_US.UTF-8'\"" $ engine-backup --mode=restore --scope=all --file=engine-20140211-1457.tar.bz2 --log=engine-backup.log --change-db-credentials --db-host=localhost --db-port=5432 --db-user=engine --db-name=engine --db-password=<PASSWORD> Restoring... Rewriting /etc/ovirt-engine/engine.conf.d/10-setup-database.conf Note: you might need to manually fix: - iptables/firewalld configuration - autostart of ovirt-engine service You can now start the engine service and then restart httpd Done.
$ engine-setup
I believe I initially created the database incorrectly (ran 'createdb engine' as postgres user). I was getting errors during engine-setup that indicated the database could not be accessed.
After the steps above, everything looks good.
- Trey
On Wed, Feb 12, 2014 at 1:10 PM, Juan Pablo Lorier <jplorier@gmail.com> wrote:
Well, too soon to say boodbye. Thought I used --change-db-credentials in the restore, the engine seems to be unable to connect to the database. I assume that it didn't get the new password, so, is there a way to tell the engine about the new password? Regards,
On 12/02/14 16:03, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=backup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=restore
Why does this work? Because 'engine-cleanup', since 3.3, does not drop the database nor user inside postgres. So when restore tries to access this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-tool, you end up almost ready, but not quite - because the password is random, and therefore different between the installations. In principle you could have provided just the password to restore, but we decided that if you need to change the credentials, you should pass all of them (except for defaults).
Hope this clarifies,
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --SnttCR8A5pcpcjpd4mLUME5X5iF7q9VSe Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable I'll do then next week. Regards and again, thanks to everyone. On 13/02/14 13:50, Trey Dockendorf wrote:
A wiki page is a good idea, glad the steps worked in more than my case.=
- Trey
On Thu, Feb 13, 2014 at 9:37 AM, Juan Pablo Lorier <jplorier@gmail.com>=
wrote:
Hi Trey,
Following your procedure, I was able to get the engine running. THANKS= A LOT TO EVERYONE!! If you allow me, I'll create a wiki page with this mentioning you so others can get this easily. Regards,
On 12/02/14 17:32, Trey Dockendorf wrote:
I was having the same issue (and posted about it today, with full steps). I'd reply to my current post to list but I don't seem to receive my own posts.
I have since taken these steps (as root)
$ su - postgres -c "dropdb engine" $ su - postgres -c "psql -c \"create user engine password '<PASSWORD>= '\"" $ su - postgres -c "psql -c \"create database engine owner engine template template0 encoding 'UTF8' lc_collate 'en_US.UTF-8' lc_ctype 'en_US.UTF-8'\"" $ engine-backup --mode=3Drestore --scope=3Dall --file=3Dengine-20140211-1457.tar.bz2 --log=3Dengine-backup.log --change-db-credentials --db-host=3Dlocalhost --db-port=3D5432 --db-user=3Dengine --db-name=3Dengine --db-password=3D<PASSWORD> Restoring... Rewriting /etc/ovirt-engine/engine.conf.d/10-setup-database.conf Note: you might need to manually fix: - iptables/firewalld configuration - autostart of ovirt-engine service You can now start the engine service and then restart httpd Done.
$ engine-setup
I believe I initially created the database incorrectly (ran 'createdb=
engine' as postgres user). I was getting errors during engine-setup that indicated the database could not be accessed.
After the steps above, everything looks good.
- Trey
Well, too soon to say boodbye. Thought I used --change-db-credentials in the restore, the engine se= ems to be unable to connect to the database. I assume that it didn't get=
On Wed, Feb 12, 2014 at 1:10 PM, Juan Pablo Lorier <jplorier@gmail.co= m> wrote: the
new password, so, is there a way to tell the engine about the new pa= ssword? Regards,
On 12/02/14 16:03, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "Sahina Bose" <sabose@redhat.com>, "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't = it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate No.
There is a specific case in which this works automatically: All on the same host: 1. engine-setup 2. engine-backup --mode=3Dbackup (perhaps do other stuff here) 3. engine-cleanup 4. engine-backup --mode=3Drestore
Why does this work? Because 'engine-cleanup', since 3.3, does not d= rop the database nor user inside postgres. So when restore tries to acc= ess this database using this user and password it succeeds.
In general, if you do the restore on another machine, and do there 'engine-setup; engine-cleanup' as a quick-postgres-provisioning-too= l, you end up almost ready, but not quite - because the password is ra= ndom, and therefore different between the installations. In principle you= could have provided just the password to restore, but we decided that if = you need to change the credentials, you should pass all of them (except= for defaults).
Hope this clarifies,
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
--SnttCR8A5pcpcjpd4mLUME5X5iF7q9VSe Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS/Ov3AAoJEC5KDDg2+NMM5ssH/jaqgEgIEwaLtwRwS3mpGnRj UO3Edpusz/Ry2V15lUjWrPBIenLwC2/Sch8jnHMb2DiLsB5pguXybT/0GUxAMDDm 3KkZt513DoG5q5Ja9EiqF148ItoW9oCQxiZQxYGKukzPUgDrMfXLlWnnKtMjTOzN 2bWkZc/Pgk/TaNU6k9nR5fPCo+wQv35OObh5qbF6K/Tr2rIyaiTvC+Hh95A6VHsJ ynqcgiF90XC3wrV0MVVjv9mBBaOhUfcg/0ulXSRjtKIwyqdkwii6GuS/ofkQx/2u OZr2Fk2YSgiOmeMgKtFZ6ERQFkuNhFlkJzmQy30CAiuejC8LQaWx27K8JbMV+VA= =Dz2U -----END PGP SIGNATURE----- --SnttCR8A5pcpcjpd4mLUME5X5iF7q9VSe--
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate Regards,
Hi Why aren't you using the --db-user --db-password parameters , its seems that you fail to authenticate
On 12/02/14 15:44, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Sahina Bose" <sabose@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:01:20 PM Subject: [Users] Problems accesing the database
Hi Sahina,
Sorry to ping you directly, but you were of great help last time. I'm also copying the list. In the process of recreating the DC I've decided to migrate the engine from a virtualbox vm to a bare metal host and I can't get engine-backup to restore the engine. Both are Centos 6.5 with engine 3.3.3, but in the new box I can't log to the database with psql -U engine -W The error is:
[root@centos-ovirt ~]# engine-backup --mode=restore --file=/root/backup_20140212.bkp --log=/root/restore.log Restoring... FATAL: Can't connect to the database
I've added to pg_hba.conf the lines I found in the web:
host all all md5
and tryied with the conf from the one that works but nothing. Any clues? Unlike engine-setup, engine-backup does nothing for you regarding creating the database - it's basically the same as creating a database to be used for a remote-db setup. See e.g. [1] for details, also 'engine-backup --help'.
[1] http://www.ovirt.org/OVirt_Engine_Development_Environment#Database
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
Hi Eli, I omitted to post that, but I did use them. It seems that others had trouble too as the database is owned by postgres after the restore and it's needed to change the engine user's password when the restore is on a different host. Thanks to all the ones that posted helpful tips, tomorrow I'll try those tips and tell the ending. Regards, El 12/02/2014 19:38, "Eli Mesika" <emesika@redhat.com> escribió:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Yedidyah Bar David" <didi@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:55:35 PM Subject: Re: [Users] Problems accesing the database
Hi Yedidyah,
But If I run engine-setup and then engine-backup restore shuldn't it import the data to the existing db created by engine-setup? That's shown everywhere so I thought it's a valid way to migrate Regards,
Hi
Why aren't you using the --db-user --db-password parameters , its seems that you fail to authenticate
On 12/02/14 15:44, Yedidyah Bar David wrote:
----- Original Message -----
From: "Juan Pablo Lorier" <jplorier@gmail.com> To: "Sahina Bose" <sabose@redhat.com> Cc: "users" <users@ovirt.org> Sent: Wednesday, February 12, 2014 7:01:20 PM Subject: [Users] Problems accesing the database
Hi Sahina,
Sorry to ping you directly, but you were of great help last time. I'm also copying the list. In the process of recreating the DC I've decided to migrate the engine from a virtualbox vm to a bare metal host and I can't get
engine-backup
to restore the engine. Both are Centos 6.5 with engine 3.3.3, but in the new box I can't log to the database with psql -U engine -W The error is:
[root@centos-ovirt ~]# engine-backup --mode=restore --file=/root/backup_20140212.bkp --log=/root/restore.log Restoring... FATAL: Can't connect to the database
I've added to pg_hba.conf the lines I found in the web:
host all all md5
and tryied with the conf from the one that works but nothing. Any clues? Unlike engine-setup, engine-backup does nothing for you regarding creating the database - it's basically the same as creating a database to be used for a remote-db setup. See e.g. [1] for details, also 'engine-backup --help'.
[1] http://www.ovirt.org/OVirt_Engine_Development_Environment#Database
_______________________________________________ Users mailing list Users@ovirt.org http://lists.ovirt.org/mailman/listinfo/users
This is an OpenPGP/MIME signed message (RFC 4880 and 3156) --gqXc6CeRIjXmLLkB2Ame5LMwJBVbqH9ja Content-Type: multipart/alternative; boundary="------------080103010100060105090101" This is a multi-part message in MIME format. --------------080103010100060105090101 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable Hi Steve, Sorry for the late reply, that link was about a different thing, I don't need to export o rebuild anything, I need to delete. Thank you very much. Regards, On 28/01/14 20:45, Steve Dainard wrote:
Not sure if this is exactly your issue, but this post here: http://comments.gmane.org/gmane.comp.emulators.ovirt.user/12200 might lead you in the right direction.
"one note - if you back it up while its attached to an engine, you will= =20 need to edit its meta data file to remove the association to allow the =
other engine to connect it to the new pool for restore."
*Steve Dainard *
On Tue, Jan 28, 2014 at 12:41 PM, Juan Pablo Lorier <jplorier@gmail.com <mailto:jplorier@gmail.com>> wrote:
Hi,
I had some issues with a gluster cluster and after some time trying= to get the storage domain up or delete it (I opened a BZ about a deadl= ock in the process of removing the domain) I gave up and destroyed the = DC. The thing is that I want to add the hosts that where part of the DC and now I get that I can't as they have the volume. I try to stop the volume but I can't as no host is running in the deleted cluster and for so= me reason, ovirt needs that. I can't delete the hosts either as they have the volume... so I'm back in another chicken and egg problem. Any hints??
PD: I can't nuke the hole ovirt plataform as I have another DC in production.... otherwise I would :-)
Regards,
_______________________________________________ Users mailing list Users@ovirt.org <mailto:Users@ovirt.org> http://lists.ovirt.org/mailman/listinfo/users
--------------080103010100060105090101 Content-Type: text/html; charset=ISO-8859-1 Content-Transfer-Encoding: quoted-printable <html> <head> <meta content=3D"text/html; charset=3DISO-8859-1" http-equiv=3D"Content-Type"> </head> <body text=3D"#000000" bgcolor=3D"#FFFFFF"> Hi Steve,<br> <br> Sorry for the late reply, that link was about a different thing, I don't need to export o rebuild anything, I need to delete.<br> Thank you very much.<br> Regards,<br> <br> <br> <div class=3D"moz-cite-prefix">On 28/01/14 20:45, Steve Dainard wrote= :<br> </div> <blockquote cite=3D"mid:CAHnsdUtE5TCQW77_yehJ1=3DVh69JH_ZamtT0DtGizyGYyn-MRqQ@mail.gm= ail.com" type=3D"cite"> <div dir=3D"ltr">Not sure if this is exactly your issue, but this post here: <a moz-do-not-send=3D"true" href=3D"http://comments.gmane.org/gmane.comp.emulators.ovirt.us= er/12200">http://comments.gmane.org/gmane.comp.emulators.ovirt.user/12200= </a> might lead you in the right direction. <div> <br> </div> <div> <pre style=3D"line-height:24px;text-align:justify">"one note - = if you back it up while its attached to an engine, you will=20 need to edit its meta data file to remove the association to allow the=20 other engine to connect it to the new pool for restore."</pre> </div> <div class=3D"gmail_extra"><br clear=3D"all"> <div> <div dir=3D"ltr"><span style=3D"font-family:arial,sans-serif;font-size:16px"><st= rong>Steve Dainard </strong></span><span style=3D"font-size:12px"></span><br> <br> </div> </div> <br> <br> <div class=3D"gmail_quote">On Tue, Jan 28, 2014 at 12:41 PM, Juan Pablo Lorier <span dir=3D"ltr"><<a moz-do-not-send=3D"true" href=3D"mailto:jplorier@gmail.co= m" target=3D"_blank" onclick=3D"window.open('https://mail.google.com/mail/?vie= w=3Dcm&tf=3D1&to=3Djplorier@gmail.com&cc=3D&bcc=3D&su= =3D&body=3D','_blank');return false;">jplorier@gmail.com</a>></span> wrote:<br> <blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi,<br> <br> I had some issues with a gluster cluster and after some time trying to<br> get the storage domain up or delete it (I opened a BZ about a deadlock<br> in the process of removing the domain) I gave up and destroyed the DC.<br> The thing is that I want to add the hosts that where part of the DC and<br> now I get that I can't as they have the volume. I try to stop the volume<br> but I can't as no host is running in the deleted cluster and for some<br> reason, ovirt needs that.<br> I can't delete the hosts either as they have the volume... so I'm back<br> in another chicken and egg problem.<br> Any hints??<br> <br> PD: I can't nuke the hole ovirt plataform as I have another DC in<br> production.... otherwise I would :-)<br> <br> Regards,<br> <br> <br> _______________________________________________<br> Users mailing list<br> <a moz-do-not-send=3D"true" href=3D"mailto:Users@ovirt.org"= onclick=3D"window.open('https://mail.google.com/mail/?vie= w=3Dcm&tf=3D1&to=3DUsers@ovirt.org&cc=3D&bcc=3D&su=3D= &body=3D','_blank');return false;">Users@ovirt.org</a><br> <a moz-do-not-send=3D"true" href=3D"http://lists.ovirt.org/mailman/listinfo/users" target=3D"_blank">http://lists.ovirt.org/mailman/listinfo= /users</a><br> <br> </blockquote> </div> <br> </div> </div> </blockquote> <br> </body> </html> --------------080103010100060105090101-- --gqXc6CeRIjXmLLkB2Ame5LMwJBVbqH9ja Content-Type: application/pgp-signature; name="signature.asc" Content-Description: OpenPGP digital signature Content-Disposition: attachment; filename="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJS6nOAAAoJEC5KDDg2+NMM1H4H/jFmU7epnjDyOh+AQ1T3mPia zx/TX+f5pj3LvYvEl3UmMRCfcPTx8zbEeuW7tv/BMsG1SC6Y1N75SYH3tIf3OgaC BSucxtzkTtYUyNUuy3Y/AaTQb2+Rb3hcSAnVekCLl5fUkdTgMpWMCFao1itRS7ye A8GRg8VBGSYsfadIiQnucbSrYnfEPxU9KqVGP9DhygRHwpp/oBJb04podUG6Jq55 JKLeYPW5Wv9m0LHcVA9tY13tnRFWyI2CJ2Vun3CcJRbp2OhgNimXfyQ5rb9zRFMO swymR/C9BNWSKuZ5r0Zr7b/xkYNYP77f7nHJ86BwePP+2hjlSo7hOV4AfGG5lhQ= =G2Yu -----END PGP SIGNATURE----- --gqXc6CeRIjXmLLkB2Ame5LMwJBVbqH9ja--
participants (8)
-
Alon Bar-Lev -
Eli Mesika -
jplorier＠gmail.com -
Juan Pablo Lorier
-
Sahina Bose -
Steve Dainard -
Trey Dockendorf -
Yedidyah Bar David