http://jenkins.ovirt.org/job/ovirt-engine_master_find-bugs_gerrit/44920/ : There was an infra issue, please contact infra(a)ovirt.org =20 From looking into the job log it appears that git failed fetching the updates from the server. That isn't the first time a git problem appears on the Jenkins CI nodes - similar failure happened on another my patch today. Is there a way to improve git communication stability on the Jenkins CI nodes?

=20 Regards, Yevgeny

_______________________________________________ Infra mailing list Infra(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/infra

-----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJXFmoBAAoJEEBxx+HSYmnD1McH/iZCw63/2D7nzlJM/tBAX80k J28afp6TcesYrF2YwlWKzin/NXND+6blDhQ0Y17/A1FAcY3caGzQT1TxVrPsRsYP Rf/szLoJn1FG4hH9HXWk7Na8bazdMNShAQuNZ9RmrQahFidoYQgVyNsZzD0BO442 WkHWeGfZhXeJ0ZshcVadGjRs8VtIw7QTJ4yTGy9YeqSCUtzUL5ljNHYI/RggVaA5 dWIKIs9mxY8t9QIUXC2EMfe7hY08qB40mZ5OCHJNscuAFVvC2nLyORjF/7TDrsY9 K65TFiKR9B1bVbuYZ7+VXzCWJ1ntdEs+fFS6MD5X6SEhi4FPGeycb9gvisSgipk= =ciSb -----END PGP SIGNATURE-----

> I'd try that approach first, though the mirror is a good idea that will > probably have to be implemented anyhow once we start adding slaves, having real > info on the network usage/errors will give us insight to actually determine > what's the issue, and thus, what's the best solution. > > @infra what do you think? > The issue with mirroring is how can you make sure that you mirror fast enough to enable CI. Even if Gerrit can push to the mirror on patch submission, there will still be some time delta between the submission happening (and the patch event showing up in Jenins) and the mirror being synced. This looks like a nasty race condition. What the mirror essentially does is make sure that bits are copied from Amazom to PHX just once. I wonder if we can get the same benefit with a simple HTTP proxy, how proxy-able is the Git HTTP protocol?

-- Barak Korren bkorren(a)redhat.com RHEV-CI Team _______________________________________________ Infra mailing list Infra(a)ovirt.org http://lists.ovirt.org/mailman/listinfo/infra

On 20 April 2016 at 10:16, Eyal Edri <eedri(a)redhat.com&gt; wrote: > > > On Wed, Apr 20, 2016 at 9:28 AM, Barak Korren <bkorren(a)redhat.com&gt; wrote: >> >> > I'd try that approach first, though the mirror is a good idea that will >> > probably have to be implemented anyhow once we start adding slaves, >> > having real >> > info on the network usage/errors will give us insight to actually >> > determine >> > what's the issue, and thus, what's the best solution. >> > >> > @infra what do you think? >> > >> >> The issue with mirroring is how can you make sure that you mirror fast >> enough to enable CI. Even if Gerrit can push to the mirror on patch >> submission, there will still be some time delta between the submission >> happening (and the patch event showing up in Jenins) and the mirror >> being synced. This looks like a nasty race condition. >> What the mirror essentially does is make sure that bits are copied >> from Amazom to PHX just once. I wonder if we can get the same benefit >> with a simple HTTP proxy, how proxy-able is the Git HTTP protocol? >> > > I think we should prioritize mirroring the GIT (not gerrit) repos to PHX, > this will help: > > Speed up all post merge jobs and reduce potential of errors from git clone > (they will be in the same network) > Reduce load (?) from the gerrit server and perhaps reduce errors of the per > patch jobs that will still run from gerrit.ovirt.org (AMAZON) > A longer goal will be either to migrate the gerrit server to PHX or to find > away to properly mirror the gerrit server (but then i fear there might be > race/problem as mentioned) > Please look at my comment about possible race conditions caused by mirroring. Simple mirroring may cause more trouble then its worth. We need to consider proxying instead.

On 04/20 10:25, Eyal Edri wrote: > On Wed, Apr 20, 2016 at 10:20 AM, Barak Korren <bkorren(a)redhat.com&gt; wro=

>=20 > > On 20 April 2016 at 10:16, Eyal Edri <eedri(a)redhat.com&gt; wrote: > > > > > > > > > On Wed, Apr 20, 2016 at 9:28 AM, Barak Korren <bkorren(a)redhat.com&gt; > > wrote: > > >> > > >> > I'd try that approach first, though the mirror is a good idea th=

> > will > > >> > probably have to be implemented anyhow once we start adding slav=

> > >> > having real > > >> > info on the network usage/errors will give us insight to actually > > >> > determine > > >> > what's the issue, and thus, what's the best solution. > > >> > > > >> > @infra what do you think? > > >> > > > >> > > >> The issue with mirroring is how can you make sure that you mirror =

> > >> enough to enable CI. Even if Gerrit can push to the mirror on patch > > >> submission, there will still be some time delta between the submis=

> > >> happening (and the patch event showing up in Jenins) and the mirror > > >> being synced. This looks like a nasty race condition. > > >> What the mirror essentially does is make sure that bits are copied > > >> from Amazom to PHX just once. I wonder if we can get the same bene=

> > >> with a simple HTTP proxy, how proxy-able is the Git HTTP protocol? > > >> > > > > > > I think we should prioritize mirroring the GIT (not gerrit) repos t=

> > > this will help: > > > > > > Speed up all post merge jobs and reduce potential of errors from git > > clone > > > (they will be in the same network) > > > Reduce load (?) from the gerrit server and perhaps reduce errors of=

> > per > > > patch jobs that will still run from gerrit.ovirt.org (AMAZON) > > > A longer goal will be either to migrate the gerrit server to PHX or=

> > find > > > away to properly mirror the gerrit server (but then i fear there mi=

> > > race/problem as mentioned) > > > > > > > Please look at my comment about possible race conditions caused by > > mirroring. Simple mirroring may cause more trouble then its worth. We > > need to consider proxying instead. > > >=20 > I don't see how a race condition can occur with a merge commit, > Can you elaborate? =20 =20 From the gerrit config on jenkins: =20 =20 Replication cache expiration time in minutes =20 If one of the server supports replication events, these events are cached=

=20 Changing this value will only take effect when Jenkins is restarted=20 =20

=20 >=20 >=20 >=20 >=20 > --=20 > Eyal Edri > Associate Manager > RHEV DevOps > EMEA ENG Virtualization R&D > Red Hat Israel >=20 > phone: +972-9-7692018 > irc: eedri (on #tlv #rhev-dev #rhev-integ) =20 --=20 David Caro =20 Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D =20 Tel.: +420 532 294 605 Email: dcaro(a)redhat.com IRC: dcaro|dcaroest@{freenode|oftc|redhat} Web: www.redhat.com RHT Global #: 82-62605

-----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQEcBAEBAgAGBQJXFzKPAAoJEEBxx+HSYmnDyEUH/016JkPNKn2PXBgXOBap2ocm tgsELdNz77v0aom34FfsB70iMYvKJNkjLYkxwTldhz4orfuL6ZWOiBRR0VuQw2lT UKUK3iUmFewB5E+q/VEUlkIJEEi99WtA/2dw5nIroc5G7SF1BirHHNpsP4BAH4bV YRX46K/4t5OYJQIjGzQEh2N+tVCpH5P5V/Dw7rjz2juefjSxVDILd3nmhag9FXQe K7ci1ORDnRFbiZKmLjaWxqBzijaZG+drxDSQU9vNcE8PY9IQPARhp2jsW/TYigIj 7qZDqqQCwUJnN5QkAWaBOSg624S8BxPwwE15+WxKJdtqk7d43163HR8/1XKDHBI= =K/Fk -----END PGP SIGNATURE-----

On 04/20 09:40, David Caro wrote: > On 04/20 10:25, Eyal Edri wrote: > > On Wed, Apr 20, 2016 at 10:20 AM, Barak Korren <bkorren(a)redhat.com&gt; wrote: > > > > > On 20 April 2016 at 10:16, Eyal Edri <eedri(a)redhat.com&gt; wrote: > > > > > > > > > > > > On Wed, Apr 20, 2016 at 9:28 AM, Barak Korren <bkorren(a)redhat.com&gt; > > > wrote: > > > >> > > > >> > I'd try that approach first, though the mirror is a good idea that > > > will > > > >> > probably have to be implemented anyhow once we start adding slaves, > > > >> > having real > > > >> > info on the network usage/errors will give us insight to actually > > > >> > determine > > > >> > what's the issue, and thus, what's the best solution. > > > >> > > > > >> > @infra what do you think? > > > >> > > > > >> > > > >> The issue with mirroring is how can you make sure that you mirror fast > > > >> enough to enable CI. Even if Gerrit can push to the mirror on patch > > > >> submission, there will still be some time delta between the submission > > > >> happening (and the patch event showing up in Jenins) and the mirror > > > >> being synced. This looks like a nasty race condition. > > > >> What the mirror essentially does is make sure that bits are copied > > > >> from Amazom to PHX just once. I wonder if we can get the same benefit > > > >> with a simple HTTP proxy, how proxy-able is the Git HTTP protocol? > > > >> > > > > > > > > I think we should prioritize mirroring the GIT (not gerrit) repos to PHX, > > > > this will help: > > > > > > > > Speed up all post merge jobs and reduce potential of errors from git > > > clone > > > > (they will be in the same network) > > > > Reduce load (?) from the gerrit server and perhaps reduce errors of the > > > per > > > > patch jobs that will still run from gerrit.ovirt.org (AMAZON) > > > > A longer goal will be either to migrate the gerrit server to PHX or to > > > find > > > > away to properly mirror the gerrit server (but then i fear there might be > > > > race/problem as mentioned) > > > > > > > > > > Please look at my comment about possible race conditions caused by > > > mirroring. Simple mirroring may cause more trouble then its worth. We > > > need to consider proxying instead. > > > > > > > I don't see how a race condition can occur with a merge commit, > > Can you elaborate? > > > From the gerrit config on jenkins: > > > Replication cache expiration time in minutes > > If one of the server supports replication events, these events are cached in memory because they can be received before the build is triggered and this plugin gets called to evaluate if the build can run. Cache allows the plugin to look if the replication events were already received when it gets called to evaluate if the build can run. If the time elapsed between this plugin gets called and the time the build entered the queue is greated than the cache expiration time, the plugin will assume that replication events were received and will let the build run. > > Changing this value will only take effect when Jenkins is restarted > And from the specific server options: Block builds in the queue until the replication events for the configured Gerrit slave(s) are received. > > > > > > > > > > > -- > > Eyal Edri > > Associate Manager > > RHEV DevOps > > EMEA ENG Virtualization R&D > > Red Hat Israel > > > > phone: +972-9-7692018 > > irc: eedri (on #tlv #rhev-dev #rhev-integ) > > -- > David Caro > > Red Hat S.L. > Continuous Integration Engineer - EMEA ENG Virtualization R&D > > Tel.: +420 532 294 605 > Email: dcaro(a)redhat.com > IRC: dcaro|dcaroest@{freenode|oftc|redhat} > Web: www.redhat.com > RHT Global #: 82-62605 -- David Caro Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D Tel.: +420 532 294 605 Email: dcaro(a)redhat.com IRC: dcaro|dcaroest@{freenode|oftc|redhat} Web: www.redhat.com RHT Global #: 82-62605

On 04/20 10:44, Eyal Edri wrote: > OK, > But if we move to use git server instead of gerrit server (only for post > merge), we won't use the Gerrit trigger plugin, so I don't see how its > still relevant. > Instead of using the gerrit trigger to run post merge jobs, we'll use the > SCM plugin instead like a normal git. > > Will this approach have any issues? Well, no feedback on gerrit, no link to the gerrit change that caused it, and only triggering on merges.

Though we already do that on some projects due to the high time they take to run. > > On Wed, Apr 20, 2016 at 10:41 AM, David Caro <dcaro(a)redhat.com&gt; wrote: > > > On 04/20 09:40, David Caro wrote: > > > On 04/20 10:25, Eyal Edri wrote: > > > > On Wed, Apr 20, 2016 at 10:20 AM, Barak Korren <bkorren(a)redhat.com > > > wrote: > > > > > > > > > On 20 April 2016 at 10:16, Eyal Edri <eedri(a)redhat.com&gt; wrote: > > > > > > > > > > > > > > > > > > On Wed, Apr 20, 2016 at 9:28 AM, Barak Korren < bkorren(a)redhat.com&gt; > > > > > wrote: > > > > > >> > > > > > >> > I'd try that approach first, though the mirror is a good idea > > that > > > > > will > > > > > >> > probably have to be implemented anyhow once we start adding > > slaves, > > > > > >> > having real > > > > > >> > info on the network usage/errors will give us insight to > > actually > > > > > >> > determine > > > > > >> > what's the issue, and thus, what's the best solution. > > > > > >> > > > > > > >> > @infra what do you think? > > > > > >> > > > > > > >> > > > > > >> The issue with mirroring is how can you make sure that you mirror > > fast > > > > > >> enough to enable CI. Even if Gerrit can push to the mirror on > > patch > > > > > >> submission, there will still be some time delta between the > > submission > > > > > >> happening (and the patch event showing up in Jenins) and the > > mirror > > > > > >> being synced. This looks like a nasty race condition. > > > > > >> What the mirror essentially does is make sure that bits are copied > > > > > >> from Amazom to PHX just once. I wonder if we can get the same > > benefit > > > > > >> with a simple HTTP proxy, how proxy-able is the Git HTTP protocol? > > > > > >> > > > > > > > > > > > > I think we should prioritize mirroring the GIT (not gerrit) repos > > to PHX, > > > > > > this will help: > > > > > > > > > > > > Speed up all post merge jobs and reduce potential of errors from > > git > > > > > clone > > > > > > (they will be in the same network) > > > > > > Reduce load (?) from the gerrit server and perhaps reduce errors > > of the > > > > > per > > > > > > patch jobs that will still run from gerrit.ovirt.org (AMAZON) > > > > > > A longer goal will be either to migrate the gerrit server to PHX > > or to > > > > > find > > > > > > away to properly mirror the gerrit server (but then i fear there > > might be > > > > > > race/problem as mentioned) > > > > > > > > > > > > > > > > Please look at my comment about possible race conditions caused by > > > > > mirroring. Simple mirroring may cause more trouble then its worth. We > > > > > need to consider proxying instead. > > > > > > > > > > > > > I don't see how a race condition can occur with a merge commit, > > > > Can you elaborate? > > > > > > > > > From the gerrit config on jenkins: > > > > > > > > > Replication cache expiration time in minutes > > > > > > If one of the server supports replication events, these events are > > cached in memory because they can be received before the build is triggered > > and this plugin gets called to evaluate if the build can run. Cache allows > > the plugin to look if the replication events were already received when it > > gets called to evaluate if the build can run. If the time elapsed between > > this plugin gets called and the time the build entered the queue is greated > > than the cache expiration time, the plugin will assume that replication > > events were received and will let the build run. > > > > > > Changing this value will only take effect when Jenkins is restarted > > > > > > > > > And from the specific server options: > > > > Block builds in the queue until the replication events for the configured > > Gerrit slave(s) are received. > > > > > > > > > > > > > > > > > > > > > > > > > -- > > > > Eyal Edri > > > > Associate Manager > > > > RHEV DevOps > > > > EMEA ENG Virtualization R&D > > > > Red Hat Israel > > > > > > > > phone: +972-9-7692018 > > > > irc: eedri (on #tlv #rhev-dev #rhev-integ) > > > > > > -- > > > David Caro > > > > > > Red Hat S.L. > > > Continuous Integration Engineer - EMEA ENG Virtualization R&D > > > > > > Tel.: +420 532 294 605 > > > Email: dcaro(a)redhat.com > > > IRC: dcaro|dcaroest@{freenode|oftc|redhat} > > > Web: www.redhat.com > > > RHT Global #: 82-62605 > > > > > > > > -- > > David Caro > > > > Red Hat S.L. > > Continuous Integration Engineer - EMEA ENG Virtualization R&D > > > > Tel.: +420 532 294 605 > > Email: dcaro(a)redhat.com > > IRC: dcaro|dcaroest@{freenode|oftc|redhat} > > Web: www.redhat.com > > RHT Global #: 82-62605 > > > > > > -- > Eyal Edri > Associate Manager > RHEV DevOps > EMEA ENG Virtualization R&D > Red Hat Israel > > phone: +972-9-7692018 > irc: eedri (on #tlv #rhev-dev #rhev-integ) -- David Caro Red Hat S.L. Continuous Integration Engineer - EMEA ENG Virtualization R&D Tel.: +420 532 294 605 Email: dcaro(a)redhat.com IRC: dcaro|dcaroest@{freenode|oftc|redhat} Web: www.redhat.com RHT Global #: 82-62605