6:28 p.m.
------=_Part_427705_1173985084.1422289723538
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Hi Punit,
You might find also tcpdump as useful tool, so then to investigate the recorded data using
Wireshark tool.
http://www.tcpdump.org/manpages/tcpdump.1.html
https://ask.wireshark.org/questions/23138/wireshark-for-red-hat-enterpris...
Thanks in advance.
Best regards,
Nikolai
____________________
Nikolai Sednev
Senior Quality Engineer at Compute team
Red Hat Israel
34 Jerusalem Road,
Ra'anana, Israel 43501
Tel: +972 9 7692043
Mobile: +972 52 7342734
Email: nsednev(a)redhat.com
IRC: nsednev
----- Original Message -----
From: users-request(a)ovirt.org
To: users(a)ovirt.org
Sent: Monday, January 26, 2015 5:27:25 PM
Subject: Users Digest, Vol 40, Issue 140
Send Users mailing list submissions to
users(a)ovirt.org
To subscribe or unsubscribe via the World Wide Web, visit
http://lists.ovirt.org/mailman/listinfo/users
or, via email, send a message with subject or body 'help' to
users-request(a)ovirt.org
You can reach the person managing the list at
users-owner(a)ovirt.org
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Users digest..."
Today's Topics:
1. Re: Update to 3.5.1 scrambled multipath.conf? (Dan Kenigsberg)
2. Re: Update to 3.5.1 scrambled multipath.conf? (Nir Soffer)
3. Re: Loopback interface has huge network transctions
(Martin Pavl?k)
4. Re: [ovirt-devel] oVirt 3.6 Feature: Cumulative Network
Usage Statistics (Dan Kenigsberg)
5. Re: Cutting edge ovirt node images. (Tolik Litovsky)
6. Re: Cutting edge ovirt node images. (Fabian Deutsch)
7. Re: Update to 3.5.1 scrambled multipath.conf? (Fabian Deutsch)
----------------------------------------------------------------------
Message: 1
Date: Mon, 26 Jan 2015 12:09:23 +0000
From: Dan Kenigsberg <danken(a)redhat.com>
To: Gianluca Cecchi <gianluca.cecchi(a)gmail.com>, nsoffer(a)redhat.com
Cc: users <users(a)ovirt.org>
Subject: Re: [ovirt-users] Update to 3.5.1 scrambled multipath.conf?
Message-ID: <20150126120923.GF14455(a)redhat.com>
Content-Type: text/plain; charset=us-ascii
On Sat, Jan 24, 2015 at 12:59:01AM +0100, Gianluca Cecchi wrote:
Hello,
on my all-in-one installation @home I had 3.5.0 with F20.
Today I updated to 3.5.1.
it seems it modified /etc/multipath.conf preventing me from using my second
disk at all...
My system has internal ssd disk (sda) for OS and one local storage domain
and another disk (sdb) with some partitions (on one of them there is also
another local storage domain).
At reboot I was put in emergency boot because partitions at sdb disk could
not be mounted (they were busy).
it took me some time to understand that the problem was due to sdb gone
managed as multipath device and so busy for partitions to be mounted.
Here you can find how multipath became after update and reboot
https://drive.google.com/file/d/0BwoPbcrMv8mvS0FkMnNyMTdVTms/view?usp=sha...
No device-mapper-multipath update in yum.log
Also it seems that after changing it, it was then reverted at boot again (I
don't know if the responsible was initrd/dracut or vdsmd) so in the mean
time the only thing I could do was to make the file immutable with
chattr +i /etc/multipath.conf
The "supported" method of achieving this is to place "# RHEV PRIVATE"
in
the second line of your hand-modified multipath.conf
I do not understand why this has happened only after upgrade to 3.5.1 -
3.5.0's should have reverted you multipath.conf just as well during each
vdsm startup.
The good thing is that this annoying behavior has been dropped from the
master branch, so that 3.6 is not going to have it. Vdsm is not to mess
with other services config file while it is running. The logic moved to
`vdsm-tool configure`
and so I was able to reboot and verify that my partitions on sdb were ok
and I was able to mount them (for safe I also ran an fsck against them)
Update ran around 19:20 and finished at 19:34
here the log in gzip format
https://drive.google.com/file/d/0BwoPbcrMv8mvWjJDTXU1YjRWOFk/view?usp=sha...
Reboot was done around 21:10-21:14
Here my /var/log/messages in gzip format, where you can see latest days.
https://drive.google.com/file/d/0BwoPbcrMv8mvMm1ldXljd3hZWnM/view?usp=sha...
Any suggestion appreciated.
Current multipath.conf (where I also commented out the getuid_callout that
is not used anymore):
[root@tekkaman setup]# cat /etc/multipath.conf
# RHEV REVISION 1.1
blacklist {
devnode "^(sda|sdb)[0-9]*"
}
defaults {
polling_interval 5
#getuid_callout "/usr/lib/udev/scsi_id --whitelisted
--replace-whitespace --device=/dev/%n"
no_path_retry fail
user_friendly_names no
flush_on_last_del yes
fast_io_fail_tmo 5
dev_loss_tmo 30
max_fds 4096
}
------------------------------
Message: 2
Date: Mon, 26 Jan 2015 07:37:59 -0500 (EST)
From: Nir Soffer <nsoffer(a)redhat.com>
To: Dan Kenigsberg <danken(a)redhat.com>
Cc: Benjamin Marzinski <bmarzins(a)redhat.com>, users <users(a)ovirt.org>
Subject: Re: [ovirt-users] Update to 3.5.1 scrambled multipath.conf?
Message-ID:
<610569491.313815.1422275879664.JavaMail.zimbra(a)redhat.com>
Content-Type: text/plain; charset=utf-8
----- Original Message -----
From: "Dan Kenigsberg" <danken(a)redhat.com>
To: "Gianluca Cecchi" <gianluca.cecchi(a)gmail.com>, nsoffer(a)redhat.com
Cc: "users" <users(a)ovirt.org>, ykaplan(a)redhat.com
Sent: Monday, January 26, 2015 2:09:23 PM
Subject: Re: [ovirt-users] Update to 3.5.1 scrambled multipath.conf?
On Sat, Jan 24, 2015 at 12:59:01AM +0100, Gianluca Cecchi wrote:
> Hello,
> on my all-in-one installation @home I had 3.5.0 with F20.
> Today I updated to 3.5.1.
>
> it seems it modified /etc/multipath.conf preventing me from using my second
> disk at all...
>
> My system has internal ssd disk (sda) for OS and one local storage domain
> and another disk (sdb) with some partitions (on one of them there is also
> another local storage domain).
>
> At reboot I was put in emergency boot because partitions at sdb disk could
> not be mounted (they were busy).
> it took me some time to understand that the problem was due to sdb gone
> managed as multipath device and so busy for partitions to be mounted.
>
> Here you can find how multipath became after update and reboot
> https://drive.google.com/file/d/0BwoPbcrMv8mvS0FkMnNyMTdVTms/view?usp=sha...
>
> No device-mapper-multipath update in yum.log
>
> Also it seems that after changing it, it was then reverted at boot again (I
> don't know if the responsible was initrd/dracut or vdsmd) so in the mean
> time the only thing I could do was to make the file immutable with
>
> chattr +i /etc/multipath.conf
The "supported" method of achieving this is to place "# RHEV PRIVATE"
in
the second line of your hand-modified multipath.conf
I do not understand why this has happened only after upgrade to 3.5.1 -
3.5.0's should have reverted you multipath.conf just as well during each
vdsm startup.
The good thing is that this annoying behavior has been dropped from the
master branch, so that 3.6 is not going to have it. Vdsm is not to mess
with other services config file while it is running. The logic moved to
`vdsm-tool configure`
>
> and so I was able to reboot and verify that my partitions on sdb were ok
> and I was able to mount them (for safe I also ran an fsck against them)
>
> Update ran around 19:20 and finished at 19:34
> here the log in gzip format
> https://drive.google.com/file/d/0BwoPbcrMv8mvWjJDTXU1YjRWOFk/view?usp=sha...
>
> Reboot was done around 21:10-21:14
>
> Here my /var/log/messages in gzip format, where you can see latest days.
> https://drive.google.com/file/d/0BwoPbcrMv8mvMm1ldXljd3hZWnM/view?usp=sha...
>
>
> Any suggestion appreciated.
>
> Current multipath.conf (where I also commented out the getuid_callout that
> is not used anymore):
>
> [root@tekkaman setup]# cat /etc/multipath.conf
> # RHEV REVISION 1.1
>
> blacklist {
> devnode "^(sda|sdb)[0-9]*"
> }
I think what happened is:
1. 3.5.1 had new multipath version
2. So vdsm upgraded the local file
3. blacklist above was removed
(it should exists in /etc/multipath.bak)
To prevent local changes, you have to mark the file as private
as Dan suggests.
Seems to be related to the find_multiapth = "yes" bug:
https://bugzilla.redhat.com/show_bug.cgi?id=1173290
Ben, can you confirm that this is the same issue?
>
> defaults {
> polling_interval 5
> #getuid_callout "/usr/lib/udev/scsi_id --whitelisted
> --replace-whitespace --device=/dev/%n"
> no_path_retry fail
> user_friendly_names no
> flush_on_last_del yes
> fast_io_fail_tmo 5
> dev_loss_tmo 30
> max_fds 4096
> }
Regards,
Nir
------------------------------
Message: 3
Date: Mon, 26 Jan 2015 14:25:23 +0100
From: Martin Pavl?k <mpavlik(a)redhat.com>
To: Punit Dambiwal <hypunit(a)gmail.com>
Cc: "users(a)ovirt.org" <users(a)ovirt.org>
Subject: Re: [ovirt-users] Loopback interface has huge network
transctions
Message-ID: <6E2536E1-27AE-47EB-BCA7-88B3901D2B02(a)redhat.com>
Content-Type: text/plain; charset="us-ascii"
Hi Punit,
it is ok since ovirt-engine is using loopback for its purposes, e.g. postgress databas
access. Try to check netstat -putna | grep 127.0.0 to see how many things are attached to
it.
If you are interested in checking what is going on a bit more have a look @ this great
how-to http://www.slashroot.in/find-network-traffic-and-bandwidth-usage-process-...
<http://www.slashroot.in/find-network-traffic-and-bandwidth-usage-process-...
HTH
Martin Pavlik
RHEV QE
On 26 Jan 2015, at 02:24, Punit Dambiwal <hypunit(a)gmail.com>
wrote:
Hi,
I have noticed that the loop back interface has huge network packets sent and
received...is it common or need to some tweaks....
1. Ovirt 3.5.1
2. Before ovirt engine installation...loop back address doesn't has that huge amount
of packets sent/receive....
3. After Ovirt engine install it's keep increasing.....and in just 48 hours it reach
to 35 GB...
[root@ccr01 ~]# ifconfig
eth0: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 43.252.x.x netmask 255.255.255.0 broadcast 43.252.x.x
ether 60:eb:69:82:0b:4c txqueuelen 1000 (Ethernet)
RX packets 6605350 bytes 6551029484 (6.1 GiB)
RX errors 0 dropped 120622 overruns 0 frame 0
TX packets 2155524 bytes 431348174 (411.3 MiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
device memory 0xdf6e0000-df700000
eth1: flags=4163<UP,BROADCAST,RUNNING,MULTICAST> mtu 1500
inet 10.10.0.2 netmask 255.255.255.0 broadcast 10.10.0.255
ether 60:eb:69:82:0b:4d txqueuelen 1000 (Ethernet)
RX packets 788160 bytes 133915292 (127.7 MiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 546352 bytes 131672255 (125.5 MiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
device memory 0xdf660000-df680000
lo: flags=73<UP,LOOPBACK,RUNNING> mtu 65536
inet 127.0.0.1 netmask 255.0.0.0
loop txqueuelen 0 (Local Loopback)
RX packets 84747311 bytes 40376482560 (37.6 GiB)
RX errors 0 dropped 0 overruns 0 frame 0
TX packets 84747311 bytes 40376482560 (37.6 GiB)
TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0
[root@ccr01 ~]# w
09:23:07 up 2 days, 11:43, 1 user, load average: 0.27, 0.30, 0.31
USER TTY LOGIN@ IDLE JCPU PCPU WHAT
root pts/0 09:16 3.00s 0.01s 0.00s w
[root@ccr01 ~]#
Thanks,
Punit
_______________________________________________
Users mailing list
Users(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/users
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.ovirt.org/pipermail/users/attachments/20150126/c38655c3/atta...
------------------------------
Message: 4
Date: Mon, 26 Jan 2015 13:45:56 +0000
From: Dan Kenigsberg <danken(a)redhat.com>
To: Lior Vernia <lvernia(a)redhat.com>
Cc: "Users(a)ovirt.org List" <Users(a)ovirt.org>, "devel(a)ovirt.org"
<devel(a)ovirt.org>
Subject: Re: [ovirt-users] [ovirt-devel] oVirt 3.6 Feature: Cumulative
Network Usage Statistics
Message-ID: <20150126134555.GI14455(a)redhat.com>
Content-Type: text/plain; charset=us-ascii
On Mon, Dec 22, 2014 at 01:40:06PM +0200, Lior Vernia wrote:
Hello users and developers,
Just put up a feature page for the aforementioned feature; in summary,
to report total RX/TX statistics for hosts and VMs in oVirt. This has
been requested several times on the users mailing list, and is
especially useful for accounting in VDI deployments.
You're more than welcome to review the feature page:
http://www.ovirt.org/Features/Cumulative_RX_TX_Statistics
Sorry for the late review; I have a couple of questions/comments.
- What do you mean by "VDI use cases" in the "Benefit to oVirt
sanpshot"
section?
Do you refer to hosting services who would like to charge their
customers based on actual bandwidth usage?
- I've added another motivation: currently-reported rxRate/txRate
can be utterly meaningless.
I don't see reference to nasty negative flows: what happens if a host
disappears? Or a VM? I suppose there's always a chance that some traffic
would go unaccounted for. But do you expect to extract this information
somehow? Either way, it should be mentioned as a caveat on the feature
page.
Note that this only deals with network usage - it'll be great if we have
similar features for CPU and disk usage!
There's a formal feature request about this:
Bug 1172153 - [RFE] Collect CPU, IO and network accounting
information
Dan
------------------------------
Message: 5
Date: Mon, 26 Jan 2015 09:00:29 -0500 (EST)
From: Tolik Litovsky <tlitovsk(a)redhat.com>
To: users(a)ovirt.org
Cc: devel(a)ovirt.org
Subject: Re: [ovirt-users] Cutting edge ovirt node images.
Message-ID: <14200256.98.1422280826405.JavaMail.tlitovsk@tolikl>
Content-Type: text/plain; charset=utf-8
Hi
Both projects are now built with VDSM and Hosted Engine plugins.
Best Regards.
Tolik.
----- Original Message -----
From: "Anatoly Litvosky" <tlitovsk(a)redhat.com>
To: users(a)ovirt.org
Cc: devel(a)ovirt.org
Sent: Thursday, 15 January, 2015 3:46:10 PM
Subject: Re: [ovirt-users] Cutting edge ovirt node images.
Hi
A centos7 based ovirt-node project joined our jenkins
http://jenkins.ovirt.org/job/ovirt-node_master_create-iso-el7_merged/
Best Regards
Tolik
On Sun, 2015-01-11 at 17:29 +0200, Anatoly Litvosky wrote:
Hi
For all of you that want the latest images for ovirt node.
For some time there is a new job in the ovirt jenkins that builds an
image for every commit made.
http://jenkins.ovirt.org/job/ovirt-node_master_create-iso-fc20_merged/
Currently for fedora 20 but hopefully more to come.
Many thanks to David (dcaro) for all the effort.
Best regards.
Tolik.
_______________________________________________
Users mailing list
Users(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/users
------------------------------
Message: 6
Date: Mon, 26 Jan 2015 10:19:29 -0500 (EST)
From: Fabian Deutsch <fdeutsch(a)redhat.com>
To: Tolik Litovsky <tlitovsk(a)redhat.com>
Cc: users(a)ovirt.org, devel(a)ovirt.org
Subject: Re: [ovirt-users] Cutting edge ovirt node images.
Message-ID:
<1747701132.306300.1422285569125.JavaMail.zimbra(a)redhat.com>
Content-Type: text/plain; charset=utf-8
Kudos Tolik!
- fabian
----- Original Message -----
Hi
Both projects are now built with VDSM and Hosted Engine plugins.
Best Regards.
Tolik.
----- Original Message -----
From: "Anatoly Litvosky" <tlitovsk(a)redhat.com>
To: users(a)ovirt.org
Cc: devel(a)ovirt.org
Sent: Thursday, 15 January, 2015 3:46:10 PM
Subject: Re: [ovirt-users] Cutting edge ovirt node images.
Hi
A centos7 based ovirt-node project joined our jenkins
http://jenkins.ovirt.org/job/ovirt-node_master_create-iso-el7_merged/
Best Regards
Tolik
On Sun, 2015-01-11 at 17:29 +0200, Anatoly Litvosky wrote:
> Hi
>
> For all of you that want the latest images for ovirt node.
> For some time there is a new job in the ovirt jenkins that builds an
> image for every commit made.
> http://jenkins.ovirt.org/job/ovirt-node_master_create-iso-fc20_merged/
> Currently for fedora 20 but hopefully more to come.
>
> Many thanks to David (dcaro) for all the effort.
>
> Best regards.
> Tolik.
>
>
>
_______________________________________________
Users mailing list
Users(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/users
_______________________________________________
Users mailing list
Users(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/users
------------------------------
Message: 7
Date: Mon, 26 Jan 2015 10:27:23 -0500 (EST)
From: Fabian Deutsch <fdeutsch(a)redhat.com>
To: Nir Soffer <nsoffer(a)redhat.com>
Cc: Benjamin Marzinski <bmarzins(a)redhat.com>, users <users(a)ovirt.org>
Subject: Re: [ovirt-users] Update to 3.5.1 scrambled multipath.conf?
Message-ID:
<1036916460.310615.1422286043504.JavaMail.zimbra(a)redhat.com>
Content-Type: text/plain; charset=utf-8
----- Original Message -----
----- Original Message -----
> From: "Dan Kenigsberg" <danken(a)redhat.com>
> To: "Gianluca Cecchi" <gianluca.cecchi(a)gmail.com>,
nsoffer(a)redhat.com
> Cc: "users" <users(a)ovirt.org>, ykaplan(a)redhat.com
> Sent: Monday, January 26, 2015 2:09:23 PM
> Subject: Re: [ovirt-users] Update to 3.5.1 scrambled multipath.conf?
>
> On Sat, Jan 24, 2015 at 12:59:01AM +0100, Gianluca Cecchi wrote:
> > Hello,
> > on my all-in-one installation @home I had 3.5.0 with F20.
> > Today I updated to 3.5.1.
> >
> > it seems it modified /etc/multipath.conf preventing me from using my
> > second
> > disk at all...
> >
> > My system has internal ssd disk (sda) for OS and one local storage domain
> > and another disk (sdb) with some partitions (on one of them there is also
> > another local storage domain).
> >
> > At reboot I was put in emergency boot because partitions at sdb disk
> > could
> > not be mounted (they were busy).
> > it took me some time to understand that the problem was due to sdb gone
> > managed as multipath device and so busy for partitions to be mounted.
> >
> > Here you can find how multipath became after update and reboot
> > https://drive.google.com/file/d/0BwoPbcrMv8mvS0FkMnNyMTdVTms/view?usp=sha...
> >
> > No device-mapper-multipath update in yum.log
> >
> > Also it seems that after changing it, it was then reverted at boot again
> > (I
> > don't know if the responsible was initrd/dracut or vdsmd) so in the mean
> > time the only thing I could do was to make the file immutable with
> >
> > chattr +i /etc/multipath.conf
>
> The "supported" method of achieving this is to place "# RHEV
PRIVATE" in
> the second line of your hand-modified multipath.conf
>
> I do not understand why this has happened only after upgrade to 3.5.1 -
> 3.5.0's should have reverted you multipath.conf just as well during each
> vdsm startup.
>
> The good thing is that this annoying behavior has been dropped from the
> master branch, so that 3.6 is not going to have it. Vdsm is not to mess
> with other services config file while it is running. The logic moved to
> `vdsm-tool configure`
>
> >
> > and so I was able to reboot and verify that my partitions on sdb were ok
> > and I was able to mount them (for safe I also ran an fsck against them)
> >
> > Update ran around 19:20 and finished at 19:34
> > here the log in gzip format
> > https://drive.google.com/file/d/0BwoPbcrMv8mvWjJDTXU1YjRWOFk/view?usp=sha...
> >
> > Reboot was done around 21:10-21:14
> >
> > Here my /var/log/messages in gzip format, where you can see latest days.
> > https://drive.google.com/file/d/0BwoPbcrMv8mvMm1ldXljd3hZWnM/view?usp=sha...
> >
> >
> > Any suggestion appreciated.
> >
> > Current multipath.conf (where I also commented out the getuid_callout
> > that
> > is not used anymore):
> >
> > [root@tekkaman setup]# cat /etc/multipath.conf
> > # RHEV REVISION 1.1
> >
> > blacklist {
> > devnode "^(sda|sdb)[0-9]*"
> > }
I think what happened is:
1. 3.5.1 had new multipath version
2. So vdsm upgraded the local file
3. blacklist above was removed
(it should exists in /etc/multipath.bak)
To prevent local changes, you have to mark the file as private
as Dan suggests.
Seems to be related to the find_multiapth = "yes" bug:
https://bugzilla.redhat.com/show_bug.cgi?id=1173290
The symptoms above sound exactly liek this issue.
When find_multipahts is no (the default when the directive is not present)
I sthat all non-blacklisted devices are tried to get claimed, and this happened above.
Blacklisting the devices works, or adding "find_mutlipaths yes" should also
work, because
in that case only device which have more than one path (or are explicitly named) will be
claimed by multipath.
My 2ct.
- fabian
Ben, can you confirm that this is the same issue?
> >
> > defaults {
> > polling_interval 5
> > #getuid_callout "/usr/lib/udev/scsi_id --whitelisted
> > --replace-whitespace --device=/dev/%n"
> > no_path_retry fail
> > user_friendly_names no
> > flush_on_last_del yes
> > fast_io_fail_tmo 5
> > dev_loss_tmo 30
> > max_fds 4096
> > }
>
Regards,
Nir
_______________________________________________
Users mailing list
Users(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/users
------------------------------
_______________________________________________
Users mailing list
Users(a)ovirt.org
http://lists.ovirt.org/mailman/listinfo/users
End of Users Digest, Vol 40, Issue 140
**************************************
------=_Part_427705_1173985084.1422289723538
Content-Type: text/html; charset=utf-8
Content-Transfer-Encoding: quoted-printable
<html><body><div style=3D"font-family: georgia,serif; font-size: 12pt;
colo=
r: #000000"><div>Hi Punit,</div><div>You might find also tcpdump
as useful =
tool, so then to investigate the recorded data using Wireshark tool.</div><=
div><a href=3D"http://www.tcpdump.org/manpages/tcpdump.1.html"
data-mce-hre=
f=3D"http://www.tcpdump.org/manpages/tcpdump.1.html">http://...
/manpages/tcpdump.1.html</a></div><div><a
href=3D"https://ask.wireshark.org=
/questions/23138/wireshark-for-red-hat-enterprise-linux" data-mce-href=3D"h=
ttps://ask.wireshark.org/questions/23138/wireshark-for-red-hat-enterprise-l=
inux">https://ask.wireshark.org/questions/23138/wireshark-for-red...
prise-linux</a></div><div><span
name=3D"x"></span><br>Thanks in advance.<br=
<div><br></div>Best
regards,<br>Nikolai<br>____________________<br>Nikolai=
Sednev<br>Senior Quality Engineer at Compute team<br>Red Hat
Israel<br>34 =
Jerusalem Road,<br>Ra'anana, Israel
43501<br><div><br></div>Tel: &nb=
sp; +972 9 7692043<br>Mobile: +972 52 7342734<br>Email:
nsedn=
ev(a)redhat.com<br>IRC: nsednev<span
name=3D"x"></span><br></div><div><br></d=
iv><hr id=3D"zwchr"><div
style=3D"color:#000;font-weight:normal;font-style:=
normal;text-decoration:none;font-family:Helvetica,Arial,sans-serif;font-siz=
e:12pt;"><b>From: </b>users-request(a)ovirt.org<br><b>To:
</b>users(a)ovirt.org=
<br><b>Sent: </b>Monday, January 26, 2015 5:27:25
PM<br><b>Subject: </b>Use=
rs Digest, Vol 40, Issue 140<br><div><br></div>Send Users mailing
list subm=
issions
to<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;users(a)ovirt.o=
rg<br><div><br></div>To subscribe or unsubscribe via the World
Wide Web, vi=
sit<br> http://lists.ovirt.o=
rg/mailman/listinfo/users<br>or, via email, send a message with subject or =
body 'help'
to<br> users-req=
uest(a)ovirt.org<br><div><br></div>You can reach the person managing
the list=
at<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;users-owner(a)ovirt.or=
g<br><div><br></div>When replying, please edit your Subject line
so it is m=
ore specific<br>than "Re: Contents of Users
digest..."<br><div><br></div><b=
r>Today's Topics:<br><div><br></div> 1.
Re: Update to 3.5=
.1 scrambled multipath.conf? (Dan Kenigsberg)<br> 2. Re:
=
Update to 3.5.1 scrambled multipath.conf? (Nir Soffer)<br> 3.
R=
e: Loopback interface has huge network transctions<br>
&n=
bsp; (Martin Pavl?k)<br> 4. Re:
[ovirt-devel] oVirt=
3.6 Feature: Cumulative Network<br> Usage
Statist=
ics (Dan Kenigsberg)<br> 5. Re: Cutting edge ovirt
node i=
mages. (Tolik Litovsky)<br> 6. Re: Cutting edge ovirt
nod=
e images. (Fabian Deutsch)<br> 7. Re: Update to 3.5.1
scr=
ambled multipath.conf? (Fabian
Deutsch)<br><div><br></div><br>-------------=
---------------------------------------------------------<br><div><br></div=
Message: 1<br>Date: Mon, 26 Jan 2015 12:09:23
+0000<br>From: Dan Kenigsber=
g &lt;danken(a)redhat.com&gt;<br>To:
Gianluca Cecchi <gianluca.cecchi@gmai=
l.com>, nsoffer(a)redhat.com<br>Cc: users
&lt;users(a)ovirt.org&gt;<br>Subje=
ct: Re: [ovirt-users] Update to 3.5.1 scrambled multipath.conf?<br>Message-=
ID: &lt;20150126120923.GF14455(a)redhat.com&gt;<br>Content-Type: text/plain;
=
charset=3Dus-ascii<br><div><br></div>On Sat, Jan 24, 2015 at
12:59:01AM +01=
00, Gianluca Cecchi wrote:<br>> Hello,<br>> on my all-in-one
installa=
tion @home I had 3.5.0 with F20.<br>> Today I updated to
3.5.1.<br>> =
<br>> it seems it modified /etc/multipath.conf preventing me from using =
my second<br>> disk at all...<br>> <br>> My system
has internal ss=
d disk (sda) for OS and one local storage domain<br>> and another disk (=
sdb) with some partitions (on one of them there is also<br>> another loc=
al storage domain).<br>> <br>> At reboot I was put in emergency
boot =
because partitions at sdb disk could<br>> not be mounted (they were busy=
).<br>> it took me some time to understand that the problem was due to s=
db gone<br>> managed as multipath device and so busy for partitions to b=
e mounted.<br>> <br>> Here you can find how multipath became
after up=
date and reboot<br>> https://drive.google.com/file/d/0BwoPbcrMv8mvS0FkMn=
NyMTdVTms/view?usp=3Dsharing<br>> <br>> No
device-mapper-multipath up=
date in yum.log<br>> <br>> Also it seems that after changing it,
it w=
as then reverted at boot again (I<br>> don't know if the responsible
was=
initrd/dracut or vdsmd) so in the mean<br>> time the only thing I could=
do was to make the file immutable with<br>> <br>> chattr +i
/etc/mul=
tipath.conf<br><div><br></div>The "supported" method of
achieving this is t=
o place "# RHEV PRIVATE" in<br>the second line of your hand-modified
multip=
ath.conf<br><div><br></div>I do not understand why this has
happened only a=
fter upgrade to 3.5.1 -<br>3.5.0's should have reverted you multipath.conf =
just as well during each<br>vdsm
startup.<br><div><br></div>The good thing =
is that this annoying behavior has been dropped from the<br>master branch, =
so that 3.6 is not going to have it. Vdsm is not to mess<br>with other serv=
ices config file while it is running. The logic moved to<br>`vdsm-tool conf=
igure`<br><div><br></div>> <br>> and so I was
able to reboot and veri=
fy that my partitions on sdb were ok<br>> and I was able to mount them (=
for safe I also ran an fsck against them)<br>> <br>> Update ran
aroun=
d 19:20 and finished at 19:34<br>> here the log in gzip
format<br>> h=
ttps://drive.google.com/file/d/0BwoPbcrMv8mvWjJDTXU1YjRWOFk/view?usp=3Dshar=
ing<br>> <br>> Reboot was done around
21:10-21:14<br>> <br>> He=
re my /var/log/messages in gzip format, where you can see latest days.<br>&=
gt; https://drive.google.com/file/d/0BwoPbcrMv8mvMm1ldXljd3hZWnM/view?usp=
=3Dsharing<br>> <br>> <br>> Any suggestion
appreciated.<br>> <b=
r>> Current multipath.conf (where I also commented out the getuid_callou=
t that<br>> is not used anymore):<br>> <br>>
[root@tekkaman setup]=
# cat /etc/multipath.conf<br>> # RHEV REVISION 1.1<br>>
<br>> blac=
klist {<br>> devnode
"^(sda|sdb)[0-9]*"<br>> }<br>> =
<br>> defaults {<br>> polling_interval
&n=
bsp; 5<br>> #getuid_callout
&nb=
sp; "/usr/lib/udev/scsi_id --whitelisted<br>>
--replace-whitespace=
--device=3D/dev/%n"<br>> no_path_retry
&nbs=
p; fail<br>> user_friendly_names
&nbs=
p; no<br>> flush_on_last_del
yes<br>&=
gt; fast_io_fail_tmo
5<br>> &nb=
sp; dev_loss_tmo
30<br>>=
max_fds
&nb=
sp; 4096<br>>
}<br><div><br></div><br>------------------------------<br>=
<div><br></div>Message: 2<br>Date: Mon, 26 Jan 2015 07:37:59 -0500
(EST)<br=
From: Nir Soffer &lt;nsoffer(a)redhat.com&gt;<br>To: Dan
Kenigsberg <dank=
en(a)redhat.com&gt;<br>Cc: Benjamin Marzinski
&lt;bmarzins(a)redhat.com&gt;, us=
ers &lt;users(a)ovirt.org&gt;<br>Subject: Re: [ovirt-users] Update to 3.5.1
s=
crambled
multipath.conf?<br>Message-ID:<br> &n=
bsp;&nbsp;&nbsp;&lt;610569491.313815.1422275879664.JavaMail.zimbra(a)redhat.c=
om><br>Content-Type: text/plain;
charset=3Dutf-8<br><div><br></div>-----=
Original Message -----<br>> From: "Dan Kenigsberg"
&lt;danken(a)redhat.co=
m><br>> To: "Gianluca Cecchi"
&lt;gianluca.cecchi(a)gmail.com&gt;, nsof=
fer(a)redhat.com<br>&gt; Cc: "users" &lt;users(a)ovirt.org&gt;,
ykaplan(a)redhat.=
com<br>> Sent: Monday, January 26, 2015 2:09:23 PM<br>> Subject:
Re: =
[ovirt-users] Update to 3.5.1 scrambled multipath.conf?<br>>
<br>> On=
Sat, Jan 24, 2015 at 12:59:01AM +0100, Gianluca Cecchi wrote:<br>>
>=
Hello,<br>> > on my all-in-one installation @home I had 3.5.0 with
F=
20.<br>> > Today I updated to 3.5.1.<br>> >
<br>> > it se=
ems it modified /etc/multipath.conf preventing me from using my second<br>&=
gt; > disk at all...<br>> > <br>> > My
system has internal s=
sd disk (sda) for OS and one local storage domain<br>> > and another
=
disk (sdb) with some partitions (on one of them there is also<br>> >
=
another local storage domain).<br>> > <br>> > At
reboot I was p=
ut in emergency boot because partitions at sdb disk could<br>> > not
=
be mounted (they were busy).<br>> > it took me some time to
understan=
d that the problem was due to sdb gone<br>> > managed as multipath
de=
vice and so busy for partitions to be mounted.<br>> >
<br>> > H=
ere you can find how multipath became after update and reboot<br>> >
=
https://drive.google.com/file/d/0BwoPbcrMv8mvS0FkMnNyMTdVTms/view?usp=3Dsha=
ring<br>> > <br>> > No device-mapper-multipath
update in yum.lo=
g<br>> > <br>> > Also it seems that after changing
it, it was t=
hen reverted at boot again (I<br>> > don't know if the responsible
wa=
s initrd/dracut or vdsmd) so in the mean<br>> > time the only thing
I=
could do was to make the file immutable with<br>> >
<br>> > ch=
attr +i /etc/multipath.conf<br>> <br>> The "supported"
method of achi=
eving this is to place "# RHEV PRIVATE" in<br>> the second line of
your =
hand-modified multipath.conf<br>> <br>> I do not understand why
this =
has happened only after upgrade to 3.5.1 -<br>> 3.5.0's should have
reve=
rted you multipath.conf just as well during each<br>> vdsm
startup.<br>&=
gt; <br>> The good thing is that this annoying behavior has been dropped=
from the<br>> master branch, so that 3.6 is not going to have it. Vdsm =
is not to mess<br>> with other services config file while it is running.=
The logic moved to<br>> `vdsm-tool configure`<br>>
<br>> > <br=
> > and so I was able to reboot and verify that my
partitions on sdb=
were ok<br>> > and I was able to mount them
(for safe I also ran an =
fsck against them)<br>> > <br>> > Update ran
around 19:20 and f=
inished at 19:34<br>> > here the log in gzip format<br>>
> http=
s://drive.google.com/file/d/0BwoPbcrMv8mvWjJDTXU1YjRWOFk/view?usp=3Dsharing=
<br>> > <br>> > Reboot was done around
21:10-21:14<br>> >=
<br>> > Here my /var/log/messages in gzip format, where you can see
=
latest days.<br>> >
https://drive.google.com/file/d/0BwoPbcrMv8mvMm1l=
dXljd3hZWnM/view?usp=3Dsharing<br>> > <br>> >
<br>> > Any=
suggestion appreciated.<br>> > <br>> > Current
multipath.conf =
(where I also commented out the getuid_callout that<br>> > is not
use=
d anymore):<br>> > <br>> > [root@tekkaman setup]#
cat /etc/mult=
ipath.conf<br>> > # RHEV REVISION 1.1<br>> >
<br>> > blac=
klist {<br>> > devnode
"^(sda|sdb)[0-9]*"<br>> > =
}<br><div><br></div><br>I think what happened
is:<br><div><br></div>1. 3.5.=
1 had new multipath version<br>2. So vdsm upgraded the local file<br>3. bla=
cklist above was removed<br> (it should exists in
/etc/multipat=
h.bak)<br><div><br></div>To prevent local changes, you have to
mark the fil=
e as private<br>as Dan suggests.<br><div><br></div>Seems to
be related to t=
he find_multiapth =3D "yes"
bug:<br>https://bugzilla.redhat.com/show_bug.cg=
i?id=3D1173290<br><div><br></div>Ben, can you confirm that this is
the same=
issue?<br><div><br></div>> > <br>>
> defaults {<br>> > &=
nbsp; polling_interval
5<br>> > &nb=
sp; #getuid_callout
"/usr/lib/udev=
/scsi_id --whitelisted<br>> > --replace-whitespace
--device=3D/dev/%n=
"<br>> > no_path_retry
&nbs=
p; fail<br>> > user_friendly_names
no<br>=
> > flush_on_last_del
yes<br>> =
> fast_io_fail_tmo
5<br>> &g=
t; dev_loss_tmo
30<b=
r>> > max_fds
=
4096<br>> > }<br>>
<br><div><br></div>Regards,<br>N=
ir<br><div><br></div><br>------------------------------<br><div><br></div>M=
essage: 3<br>Date: Mon, 26 Jan 2015 14:25:23 +0100<br>From: Martin Pavl?k
&=
lt;mpavlik(a)redhat.com&gt;<br>To: Punit Dambiwal
&lt;hypunit(a)gmail.com&gt;<b=
r>Cc: "users(a)ovirt.org" &lt;users(a)ovirt.org&gt;<br>Subject: Re:
[ovirt-user=
s] Loopback interface has huge
network<br> &nb=
sp; transctions<br>Message-ID:
<6E2536E1-27AE-47EB-BCA7-88B39=
01D2B02(a)redhat.com&gt;<br>Content-Type: text/plain;
charset=3D"us-ascii"<br=
<div><br></div>Hi
Punit,<br><div><br></div>it is ok since ovirt-engine is =
using loopback for its purposes, e.g. postgress databas access. Try to chec=
k netstat -putna | grep 127.0.0 to see how many things are attached to it.<=
br><div><br></div>If you are interested in checking what is going on
a bit =
more have a look @ this great how-to http://www.slashroot.in/find-network-t=
raffic-and-bandwidth-usage-process-linux <http://www.slashroot.in/find-n=
etwork-traffic-and-bandwidth-usage-process-linux><br><div><br></div><br>=
HTH <br><div><br></div>Martin
Pavlik<br><div><br></div>RHEV QE<br><div><br>=
</div>> On 26 Jan 2015, at 02:24, Punit Dambiwal
&lt;hypunit(a)gmail.com&g=
t; wrote:<br>> <br>> Hi,<br>> <br>> I
have noticed that the loo=
p back interface has huge network packets sent and received...is it common =
or need to some tweaks.... <br>> <br>> 1. Ovirt
3.5.1<br>> 2. Befo=
re ovirt engine installation...loop back address doesn't has that huge amou=
nt of packets sent/receive....<br>> 3. After Ovirt engine install it's
k=
eep increasing.....and in just 48 hours it reach to 35 GB...<br>>
<br>&g=
t; [root@ccr01 ~]# ifconfig<br>> eth0:
flags=3D4163<UP,BROADCAST,RUNN=
ING,MULTICAST> mtu 1500<br>>
inet 4=
3.252.x.x netmask 255.255.255.0 broadcast 43.252.x.x<br>>
&n=
bsp; ether 60:eb:69:82:0b:4c txqueuelen 1000
&nb=
sp;(Ethernet)<br>> RX packets
6605350 =
bytes 6551029484 (6.1 GiB)<br>>
RX errors 0 =
dropped 120622 overruns 0 frame 0<br>>
&=
nbsp; TX packets 2155524 bytes 431348174 (411.3
MiB)<br>> &=
nbsp; TX errors 0 dropped 0 overruns 0
car=
rier 0 collisions 0<br>>
device memory=
0xdf6e0000-df700000<br>> <br>> eth1:
flags=3D4163<UP,BROADCAST,RU=
NNING,MULTICAST> mtu 1500<br>>
inet=
10.10.0.2 netmask 255.255.255.0 broadcast
10.10.0.255<br>> =
ether 60:eb:69:82:0b:4d txqueuelen
1000 &=
nbsp;(Ethernet)<br>> RX packets
788160  =
;bytes 133915292 (127.7 MiB)<br>>
RX errors =
0 dropped 0 overruns 0 frame 0<br>>
&nbs=
p; TX packets 546352 bytes 131672255 (125.5 MiB)<br>>
 =
; TX errors 0 dropped 0 overruns 0
carrier=
0 collisions 0<br>>
device memory 0xd=
f660000-df680000<br>> <br>> lo:
flags=3D73<UP,LOOPBACK,RUNNING>=
mtu 65536<br>> inet
127.0.0.1 n=
etmask 255.0.0.0<br>> loop
txqueuelen =
0 (Local Loopback)<br>>
RX packets 847=
47311 bytes 40376482560 (37.6 GiB)<br>>
 =
; RX errors 0 dropped 0 overruns 0 frame 0<br>>
=
TX packets 84747311 bytes 40376482560 (37.6
GiB=
)<br>> TX errors 0
dropped 0 overruns =
0 carrier 0 collisions 0<br>> <br>>
[root@ccr01 ~]# w<br>=
> 09:23:07 up 2 days, 11:43, 1 user, load average:
0.2=
7, 0.30, 0.31<br>> USER TTY
LOG=
IN@ IDLE JCPU PCPU WHAT<br>> root
pts=
/0 09:16 3.00s 0.01s 0.00s
w<br>>=
[root@ccr01 ~]#<br>> <br>> Thanks,<br>>
Punit<br>> ___________=
____________________________________<br>> Users mailing
list<br>> Use=
rs(a)ovirt.org<br>&gt;
http://lists.ovirt.org/mailman/listinfo/users<br><div>=
<br></div>-------------- next part --------------<br>An HTML attachment
was=
scrubbed...<br>URL: <http://lists.ovirt.org/pipermail/users/attachments=
/20150126/c38655c3/attachment-0001.html><br><div><br></div>-------------=
-----------------<br><div><br></div>Message: 4<br>Date: Mon,
26 Jan 2015 13=
:45:56 +0000<br>From: Dan Kenigsberg &lt;danken(a)redhat.com&gt;<br>To:
Lior =
Vernia &lt;lvernia(a)redhat.com&gt;<br>Cc: "Users(a)ovirt.org List"
<Users@o=
virt.org&gt;,&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;"devel(a)ovirt.o=
rg"<br>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&lt;devel(a)ovirt.org&=
gt;<br>Subject: Re: [ovirt-users] [ovirt-devel] oVirt 3.6 Feature: Cumulati=
ve<br> Network
Usage Statist=
ics<br>Message-ID:
&lt;20150126134555.GI14455(a)redhat.com&gt;<br>Content-Typ=
e: text/plain; charset=3Dus-ascii<br><div><br></div>On Mon, Dec
22, 2014 at=
01:40:06PM +0200, Lior Vernia wrote:<br>> Hello users and developers,<b=
r>> <br>> Just put up a feature page for the aforementioned
feature; =
in summary,<br>> to report total RX/TX statistics for hosts and VMs in o=
Virt. This has<br>> been requested several times on the users mailing li=
st, and is<br>> especially useful for accounting in VDI
deployments.<br>=
> <br>> You're more than welcome to review the feature
page:<br>> =
http://www.ovirt.org/Features/Cumulative_RX_TX_Statistics<br><di...
Sorry for the late review; I have a couple of
questions/comments.<br>- Wha=
t do you mean by "VDI use cases" in
the "Benefit to oVirt sanpshot"<br>&nbs=
p; section?<br> Do you refer to hosting services who
would =
like to charge their<br> customers based on actual bandwidth
usa=
ge?<br>- I've added another motivation: currently-reported rxRate/txRate<br=
can be utterly
meaningless.<br><div><br></div><br>I don't see =
reference to nasty negative flows: what happens if a host<br>disappears? Or=
a VM? I suppose there's always a chance that some traffic<br>would go unac=
counted for. But do you expect to extract this information<br>somehow? Eith=
er way, it should be mentioned as a caveat on the
feature<br>page.<br><div>=
<br></div>> <br>> Note that this only deals with network
usage - it'l=
l be great if we have<br>> similar features for CPU and disk
usage!<br><=
div><br></div>There's a formal feature request about
this:<br> =
Bug 1172153 - [RFE] Collect CPU, IO and network
accounting<br> &=
nbsp;
information<br><div><br></div>Dan<br><div><br></div><br>-------=
-----------------------<br><div><br></div>Message:
5<br>Date: Mon, 26 Jan 2=
015 09:00:29 -0500 (EST)<br>From: Tolik Litovsky
&lt;tlitovsk(a)redhat.com&gt=
;<br>To: users(a)ovirt.org<br>Cc: devel(a)ovirt.org<br>Subject: Re:
[ovirt-user=
s] Cutting edge ovirt node images.<br>Message-ID: <14200256.98.142228082=
6405.JavaMail.tlitovsk@tolikl><br>Content-Type: text/plain; charset=3Dut=
f-8<br><div><br></div>Hi<br><div><br></div>Both
projects are now built with=
VDSM and Hosted Engine plugins.<br><div><br></div>Best
Regards.<br>Tolik.<=
br><div><br></div>----- Original Message -----<br>From:
"Anatoly Litvosky" =
&lt;tlitovsk(a)redhat.com&gt;<br>To: users(a)ovirt.org<br>Cc:
devel(a)ovirt.org<b=
r>Sent: Thursday, 15 January, 2015 3:46:10 PM<br>Subject: Re: [ovirt-users]=
Cutting edge ovirt node images.<br><div><br></div>Hi
<br><div><br></div>A =
centos7 based ovirt-node project joined our jenkins<br>http://jenkins.ovirt=
.org/job/ovirt-node_master_create-iso-el7_merged/<br><div><br></div>Best
Re=
gards<br>Tolik<br><div><br></div>On Sun, 2015-01-11 at 17:29
+0200, Anatoly=
Litvosky wrote:<br>> Hi<br>> <br>> For all of you
that want the l=
atest images for ovirt node.<br>> For some time there is a new job in th=
e ovirt jenkins that builds an<br>> image for every commit
made.<br>>=
http://jenkins.ovirt.org/job/ovirt-node_master_create-iso-fc20_merged/<...
> Currently for fedora 20 but hopefully more to come.<br>>
<br>> M=
any thanks to David (dcaro) for all the effort.<br>> <br>> Best
regar=
ds.<br>> Tolik.<br>> <br>> <br>>
<br><div><br></div><br>_______=
________________________________________<br>Users mailing list<br>Users@ovi=
rt.org<br>http://lists.ovirt.org/mailman/listinfo/users<br>&l...
br>------------------------------<br><div><br></div>Message:
6<br>Date: Mon=
, 26 Jan 2015 10:19:29 -0500 (EST)<br>From: Fabian Deutsch <fdeutsch@red=
hat.com><br>To: Tolik Litovsky
&lt;tlitovsk(a)redhat.com&gt;<br>Cc: users@=
ovirt.org, devel(a)ovirt.org<br>Subject: Re: [ovirt-users] Cutting edge ovirt=
node
images.<br>Message-ID:<br> &=
nbsp;&lt;1747701132.306300.1422285569125.JavaMail.zimbra(a)redhat.com&gt;<br>=
Content-Type: text/plain; charset=3Dutf-8<br><div><br></div>Kudos
Tolik!<br=
<div><br></div>-
fabian<br><div><br></div>----- Original Message -----<br>=
> Hi<br>> <br>> Both projects are now built with
VDSM and Hosted E=
ngine plugins.<br>> <br>> Best Regards.<br>>
Tolik.<br>> <br>&g=
t; ----- Original Message -----<br>> From: "Anatoly Litvosky"
<tlitov=
sk(a)redhat.com&gt;<br>&gt; To: users(a)ovirt.org<br>&gt; Cc:
devel(a)ovirt.org<b=
r>> Sent: Thursday, 15 January, 2015 3:46:10 PM<br>> Subject: Re:
[ov=
irt-users] Cutting edge ovirt node images.<br>> <br>>
Hi<br>> <br>=
> A centos7 based ovirt-node project joined our jenkins<br>>
http://j=
enkins.ovirt.org/job/ovirt-node_master_create-iso-el7_merged/<br>&a...
<br>&=
gt; Best Regards<br>> Tolik<br>> <br>> On Sun,
2015-01-11 at 17:29=
+0200, Anatoly Litvosky wrote:<br>> > Hi<br>> >
<br>> > =
For all of you that want the latest images for ovirt node.<br>> >
For=
some time there is a new job in the ovirt jenkins that builds an<br>>
&=
gt; image for every commit made.<br>> >
http://jenkins.ovirt.org/job/=
ovirt-node_master_create-iso-fc20_merged/<br>> > Currently for
fedora=
20 but hopefully more to come.<br>> > <br>> >
Many thanks to D=
avid (dcaro) for all the effort.<br>> > <br>> >
Best regards.<b=
r>> > Tolik.<br>> > <br>> >
<br>> > <br>> <br>&g=
t; <br>> _______________________________________________<br>>
Users m=
ailing list<br>> Users(a)ovirt.org<br>&gt;
http://lists.ovirt.org/mailman/=
listinfo/users<br>>
_______________________________________________<br>&=
gt; Users mailing list<br>> Users(a)ovirt.org<br>&gt;
http://lists.ovirt.o=
rg/mailman/listinfo/users<br>>
<br><div><br></div><br>------------------=
------------<br><div><br></div>Message: 7<br>Date: Mon, 26
Jan 2015 10:27:2=
3 -0500 (EST)<br>From: Fabian Deutsch
&lt;fdeutsch(a)redhat.com&gt;<br>To: Ni=
r Soffer &lt;nsoffer(a)redhat.com&gt;<br>Cc: Benjamin Marzinski
<bmarzins@=
redhat.com>, users &lt;users(a)ovirt.org&gt;<br>Subject: Re:
[ovirt-users]=
Update to 3.5.1 scrambled
multipath.conf?<br>Message-ID:<br> &n=
bsp; <1036916460.310615.1422286043504.JavaM=
ail.zimbra(a)redhat.com&gt;<br>Content-Type: text/plain;
charset=3Dutf-8<br><=
div><br></div><br><div><br></div>----- Original
Message -----<br>> -----=
Original Message -----<br>> > From: "Dan Kenigsberg"
<danken@redh=
at.com><br>> > To: "Gianluca Cecchi"
&lt;gianluca.cecchi(a)gmail.com=
>, nsoffer(a)redhat.com<br>&gt; > Cc: "users"
&lt;users(a)ovirt.org&gt;, =
ykaplan(a)redhat.com<br>&gt; > Sent: Monday, January 26, 2015 2:09:23
PM<b=
r>> > Subject: Re: [ovirt-users] Update to 3.5.1 scrambled multipath.=
conf?<br>> > <br>> > On Sat, Jan 24, 2015 at
12:59:01AM +0100, =
Gianluca Cecchi wrote:<br>> > > Hello,<br>>
> > on my all=
-in-one installation @home I had 3.5.0 with F20.<br>> > > Today
I =
updated to 3.5.1.<br>> > > <br>> > >
it seems it modified=
/etc/multipath.conf preventing me from using my<br>> > >
second<b=
r>> > > disk at all...<br>> > >
<br>> > > My sys=
tem has internal ssd disk (sda) for OS and one local storage domain<br>>=
> > and another disk (sdb) with some partitions (on one of them ther=
e is also<br>> > > another local storage
domain).<br>> > >=
; <br>> > > At reboot I was put in emergency boot because
partitio=
ns at sdb disk<br>> > > could<br>> >
> not be mounted (th=
ey were busy).<br>> > > it took me some time to understand that
th=
e problem was due to sdb gone<br>> > > managed as multipath
device=
and so busy for partitions to be mounted.<br>> > >
<br>> > =
> Here you can find how multipath became after update and
reboot<br>>=
> > https://drive.google.com/file/d/0BwoPbcrMv8mvS0FkMnNyMTdVTms/vie=
w?usp=3Dsharing<br>> > > <br>> > >
No device-mapper-multi=
path update in yum.log<br>> > > <br>> >
> Also it seems t=
hat after changing it, it was then reverted at boot again<br>> >
>=
(I<br>> > > don't know if the responsible was
initrd/dracut or vd=
smd) so in the mean<br>> > > time the only thing I could do was
to=
make the file immutable with<br>> > > <br>>
> > chattr +=
i /etc/multipath.conf<br>> > <br>> > The
"supported" method of =
achieving this is to place "# RHEV PRIVATE" in<br>> > the
second line=
of your hand-modified multipath.conf<br>> > <br>>
> I do not u=
nderstand why this has happened only after upgrade to 3.5.1 -<br>> >
=
3.5.0's should have reverted you multipath.conf just as well during each<br=
> > vdsm startup.<br>> >
<br>> > The good thing is that =
this annoying behavior has been
dropped from the<br>> > master branch=
, so that 3.6 is not going to have it. Vdsm is not to mess<br>> >
wit=
h other services config file while it is running. The logic moved to<br>>=
; > `vdsm-tool configure`<br>> > <br>> >
> <br>> > =
> and so I was able to reboot and verify that my partitions on sdb were =
ok<br>> > > and I was able to mount them (for safe I also ran
an f=
sck against them)<br>> > > <br>> > >
Update ran around 19=
:20 and finished at 19:34<br>> > > here the log in gzip
format<br>=
> > > https://drive.google.com/file/d/0BwoPbcrMv8mvWjJDTXU1YjRWOFk=
/view?usp=3Dsharing<br>> > > <br>> >
> Reboot was done ar=
ound 21:10-21:14<br>> > > <br>> > >
Here my /var/log/mess=
ages in gzip format, where you can see latest days.<br>> > >
https=
://drive.google.com/file/d/0BwoPbcrMv8mvMm1ldXljd3hZWnM/view?usp=3Dsharing<=
br>> > > <br>> > > <br>>
> > Any suggestion appr=
eciated.<br>> > > <br>> > > Current
multipath.conf (where=
I also commented out the getuid_callout<br>> > >
that<br>> >=
; > is not used anymore):<br>> > > <br>>
> > [root@tek=
kaman setup]# cat /etc/multipath.conf<br>> > > # RHEV REVISION
1.1=
<br>> > > <br>> > > blacklist
{<br>> > > =
devnode "^(sda|sdb)[0-9]*"<br>> > >
}<br>> <br>> <br>=
> I think what happened is:<br>> <br>> 1. 3.5.1 had new
multipath =
version<br>> 2. So vdsm upgraded the local file<br>> 3.
blacklist abo=
ve was removed<br>> (it should exists in
/etc/multipath.bak=
)<br>> <br>> To prevent local changes, you have to mark the file
as p=
rivate<br>> as Dan suggests.<br>> <br>> Seems to be
related to the=
find_multiapth =3D "yes" bug:<br>>
https://bugzilla.redhat.com/show_bug=
.cgi?id=3D1173290<br><div><br></div>The symptoms above sound
exactly liek t=
his issue.<br>When find_multipahts is no (the default when the directive is=
not present)<br>I sthat all non-blacklisted devices are tried to get claim=
ed, and this happened above.<br><div><br></div>Blacklisting the
devices wor=
ks, or adding "find_mutlipaths yes" should also work, because<br>in that
ca=
se only device which have more than one path (or are explicitly named) will=
be<br>claimed by multipath.<br><div><br></div>My
2ct.<br><div><br></div>- =
fabian<br><div><br></div>> Ben, can you confirm that this
is the same is=
sue?<br>> <br>> > > <br>> >
> defaults {<br>> > =
> polling_interval
5<br>> &g=
t; > #getuid_callout
"/u=
sr/lib/udev/scsi_id --whitelisted<br>> > > --replace-whitespace
--=
device=3D/dev/%n"<br>> > >
no_path_retry &nbs=
p; fail<br>> > >
user_friendly_=
names no<br>> > >
flush_on_last_del &n=
bsp; yes<br>> > >
fast_io_fail_tmo &nb=
sp; 5<br>> > >
dev_loss_tmo &nbs=
p; 30<br>> >
> max=
_fds
4096<br>> &=
gt; > }<br>> > <br>> <br>>
Regards,<br>> Nir<br>> ____=
___________________________________________<br>> Users mailing
list<br>&=
gt; Users(a)ovirt.org<br>&gt; http://lists.ovirt.org/mailman/listinfo/users<b=
r>>
<br><div><br></div><br>------------------------------<br><div><br></=
div>_______________________________________________<br>Users mailing list<b=
r>Users@ovirt.org<br>http://lists.ovirt.org/mailman/listinfo/users<br><div>=
<br></div><br>End of Users Digest, Vol 40, Issue
140<br>*******************=
*******************<br></div><div><br></div></div></body></html>
------=_Part_427705_1173985084.1422289723538--
3589
days inactive
3589
days old
0 comments
1 participants
participants (1)
-
Nikolai Sednev