From florin.papa at intel.com Mon Nov 2 08:07:19 2015 From: florin.papa at intel.com (Papa, Florin) Date: Mon, 2 Nov 2015 13:07:19 +0000 Subject: [Python-buildbots] SkyLake Intel Buildbot Message-ID: <3A375A669FBEFF45B6B60E689636EDCA09B534F0@IRSMSX101.ger.corp.intel.com> Hi all, This is Florin Papa from the Intel Corporation. I would like to set up a Python buildbot on a SkyLake computer in order to have hardware support for the MPX feature proposed here: http://bugs.python.org/issue25300. Please provide me with a suitable name and password. Thank you, Florin Papa From zachary.ware+pydev at gmail.com Mon Nov 2 20:50:48 2015 From: zachary.ware+pydev at gmail.com (Zachary Ware) Date: Mon, 2 Nov 2015 19:50:48 -0600 Subject: [Python-buildbots] SkyLake Intel Buildbot In-Reply-To: <3A375A669FBEFF45B6B60E689636EDCA09B534F0@IRSMSX101.ger.corp.intel.com> References: <3A375A669FBEFF45B6B60E689636EDCA09B534F0@IRSMSX101.ger.corp.intel.com> Message-ID: Hi Florin, On Mon, Nov 2, 2015 at 7:07 AM, Papa, Florin wrote: > Hi all, > > This is Florin Papa from the Intel Corporation. I would like to set up a Python buildbot on a SkyLake computer in order to have hardware support for the MPX feature proposed here: http://bugs.python.org/issue25300. We generally form the slave name from the name of the owner, the operating system, and possibly the architecture. What are those values for your slave? Once we have that information, I'll add the slave to the master and send you the name/password off-list. -- Zach From florin.papa at intel.com Tue Nov 3 02:54:01 2015 From: florin.papa at intel.com (Papa, Florin) Date: Tue, 3 Nov 2015 07:54:01 +0000 Subject: [Python-buildbots] SkyLake Intel Buildbot In-Reply-To: References: <3A375A669FBEFF45B6B60E689636EDCA09B534F0@IRSMSX101.ger.corp.intel.com> Message-ID: <3A375A669FBEFF45B6B60E689636EDCA09B53612@IRSMSX101.ger.corp.intel.com> Hi Zach, The values for the requested fields are: Owner: Intel OS: Ubuntu 15.10 64-bit Architecture: Intel Core i7-6700K aka Skylake Thank you, Florin -----Original Message----- From: zachary.ware at gmail.com [mailto:zachary.ware at gmail.com] On Behalf Of Zachary Ware Sent: Tuesday, November 3, 2015 3:51 AM To: Papa, Florin Cc: python-buildbots at python.org; Popa, Stefan A ; Manciu, Catalin Gabriel ; Patrascu, Alecsandru Subject: Re: [Python-buildbots] SkyLake Intel Buildbot Hi Florin, On Mon, Nov 2, 2015 at 7:07 AM, Papa, Florin wrote: > Hi all, > > This is Florin Papa from the Intel Corporation. I would like to set up a Python buildbot on a SkyLake computer in order to have hardware support for the MPX feature proposed here: http://bugs.python.org/issue25300. We generally form the slave name from the name of the owner, the operating system, and possibly the architecture. What are those values for your slave? Once we have that information, I'll add the slave to the master and send you the name/password off-list. -- Zach From rdmurray at bitdance.com Tue Nov 3 15:23:32 2015 From: rdmurray at bitdance.com (R. David Murray) Date: Tue, 03 Nov 2015 15:23:32 -0500 Subject: [Python-buildbots] SkyLake Intel Buildbot In-Reply-To: <3A375A669FBEFF45B6B60E689636EDCA09B53612@IRSMSX101.ger.corp.intel.com> References: <3A375A669FBEFF45B6B60E689636EDCA09B534F0@IRSMSX101.ger.corp.intel.com> <3A375A669FBEFF45B6B60E689636EDCA09B53612@IRSMSX101.ger.corp.intel.com> Message-ID: <20151103202333.D3F93B14095@webabinitio.net> On Tue, 03 Nov 2015 07:54:01 +0000, "Papa, Florin" wrote: > The values for the requested fields are: > Owner: Intel > OS: Ubuntu 15.10 64-bit > Architecture: Intel Core i7-6700K aka Skylake The skylake is the unique aspect of this buildbot, so either that or the i7-6700K should probably be in the name somewhere. --David From florin.papa at intel.com Wed Nov 4 10:33:17 2015 From: florin.papa at intel.com (Papa, Florin) Date: Wed, 4 Nov 2015 15:33:17 +0000 Subject: [Python-buildbots] SkyLake Intel Buildbot In-Reply-To: <20151103202333.D3F93B14095@webabinitio.net> References: <3A375A669FBEFF45B6B60E689636EDCA09B534F0@IRSMSX101.ger.corp.intel.com> <3A375A669FBEFF45B6B60E689636EDCA09B53612@IRSMSX101.ger.corp.intel.com> <20151103202333.D3F93B14095@webabinitio.net> Message-ID: <3A375A669FBEFF45B6B60E689636EDCA09B5385D@IRSMSX101.ger.corp.intel.com> I think "Skylake" works fine. Thank you, Florin -----Original Message----- From: R. David Murray [mailto:rdmurray at bitdance.com] Sent: Tuesday, November 3, 2015 10:24 PM To: Papa, Florin Cc: Popa, Stefan A ; Manciu, Catalin Gabriel ; python-buildbots at python.org; Patrascu, Alecsandru Subject: Re: [Python-buildbots] SkyLake Intel Buildbot On Tue, 03 Nov 2015 07:54:01 +0000, "Papa, Florin" wrote: > The values for the requested fields are: > Owner: Intel > OS: Ubuntu 15.10 64-bit > Architecture: Intel Core i7-6700K aka Skylake The skylake is the unique aspect of this buildbot, so either that or the i7-6700K should probably be in the name somewhere. --David From zachary.ware+pydev at gmail.com Thu Nov 5 14:24:00 2015 From: zachary.ware+pydev at gmail.com (Zachary Ware) Date: Thu, 5 Nov 2015 13:24:00 -0600 Subject: [Python-buildbots] SkyLake Intel Buildbot In-Reply-To: <20151103202333.D3F93B14095@webabinitio.net> References: <3A375A669FBEFF45B6B60E689636EDCA09B534F0@IRSMSX101.ger.corp.intel.com> <3A375A669FBEFF45B6B60E689636EDCA09B53612@IRSMSX101.ger.corp.intel.com> <20151103202333.D3F93B14095@webabinitio.net> Message-ID: On Tue, Nov 3, 2015 at 2:23 PM, R. David Murray wrote: > On Tue, 03 Nov 2015 07:54:01 +0000, "Papa, Florin" wrote: >> The values for the requested fields are: >> Owner: Intel >> OS: Ubuntu 15.10 64-bit >> Architecture: Intel Core i7-6700K aka Skylake > > The skylake is the unique aspect of this buildbot, so either that or > the i7-6700K should probably be in the name somewhere. > > --David 'intel-ubuntu-skylake' is now set up on the master, password sent privately. Currently it's just set up to do a standard Unix build (./configure --with-pydebug, make, make buildbottest, make clean); we'll adjust that when we have something to adjust it to. -- Zach From florin.papa at intel.com Fri Nov 6 10:43:24 2015 From: florin.papa at intel.com (Papa, Florin) Date: Fri, 6 Nov 2015 15:43:24 +0000 Subject: [Python-buildbots] SkyLake Intel Buildbot In-Reply-To: References: <3A375A669FBEFF45B6B60E689636EDCA09B534F0@IRSMSX101.ger.corp.intel.com> <3A375A669FBEFF45B6B60E689636EDCA09B53612@IRSMSX101.ger.corp.intel.com> <20151103202333.D3F93B14095@webabinitio.net> Message-ID: <3A375A669FBEFF45B6B60E689636EDCA09B5B98F@IRSMSX101.ger.corp.intel.com> The Skylake buildbot is now set, it appears as idle here: http://buildbot.python.org/all/waterfall Regards, Florin -----Original Message----- From: zachary.ware at gmail.com [mailto:zachary.ware at gmail.com] On Behalf Of Zachary Ware Sent: Thursday, November 5, 2015 9:24 PM To: python-buildbots at python.org Cc: Papa, Florin ; Popa, Stefan A ; Manciu, Catalin Gabriel ; Patrascu, Alecsandru Subject: Re: [Python-buildbots] SkyLake Intel Buildbot On Tue, Nov 3, 2015 at 2:23 PM, R. David Murray wrote: > On Tue, 03 Nov 2015 07:54:01 +0000, "Papa, Florin" wrote: >> The values for the requested fields are: >> Owner: Intel >> OS: Ubuntu 15.10 64-bit >> Architecture: Intel Core i7-6700K aka Skylake > > The skylake is the unique aspect of this buildbot, so either that or > the i7-6700K should probably be in the name somewhere. > > --David 'intel-ubuntu-skylake' is now set up on the master, password sent privately. Currently it's just set up to do a standard Unix build (./configure --with-pydebug, make, make buildbottest, make clean); we'll adjust that when we have something to adjust it to. -- Zach From dje.gcc at gmail.com Mon Nov 16 13:11:44 2015 From: dje.gcc at gmail.com (David Edelsohn) Date: Mon, 16 Nov 2015 13:11:44 -0500 Subject: [Python-buildbots] Stable builders Message-ID: What's the plan about updating the list of stable builders? The current stable builders don't appear very stable (not all green). There are other builders that are stable and should be added to the stable builders list. Thanks, David From rdmurray at bitdance.com Mon Nov 16 13:35:31 2015 From: rdmurray at bitdance.com (R. David Murray) Date: Mon, 16 Nov 2015 13:35:31 -0500 Subject: [Python-buildbots] Stable builders In-Reply-To: References: Message-ID: <20151116183531.90AB4B14091@webabinitio.net> On Mon, 16 Nov 2015 13:11:44 -0500, David Edelsohn wrote: > What's the plan about updating the list of stable builders? > > The current stable builders don't appear very stable (not all green). At a quick glance it looks like those are bugs that need fixing, not stability errors. We should ping the owners of the ones that are offline, though. > There are other builders that are stable and should be added to the > stable builders list. Do you have ones you'd like to nominate? It's probably time to put the docs one into stable. Unless there are outstanding issues with it, Zach? The ICC mac and unix buildbots are stable, but we can't put those into the stable set without talking to python-dev. It seems like it is time to do that. --David From dje.gcc at gmail.com Mon Nov 16 13:54:21 2015 From: dje.gcc at gmail.com (David Edelsohn) Date: Mon, 16 Nov 2015 13:54:21 -0500 Subject: [Python-buildbots] Stable builders In-Reply-To: <20151116183531.90AB4B14091@webabinitio.net> References: <20151116183531.90AB4B14091@webabinitio.net> Message-ID: On Mon, Nov 16, 2015 at 1:35 PM, R. David Murray wrote: > On Mon, 16 Nov 2015 13:11:44 -0500, David Edelsohn wrote: >> There are other builders that are stable and should be added to the >> stable builders list. > > Do you have ones you'd like to nominate? > > It's probably time to put the docs one into stable. Unless there are > outstanding issues with it, Zach? I would like to propose the PPC64LE builder and one or all of the s390x builders. Thanks, David From rosuav at gmail.com Mon Nov 16 15:38:11 2015 From: rosuav at gmail.com (Chris Angelico) Date: Tue, 17 Nov 2015 07:38:11 +1100 Subject: [Python-buildbots] Stable builders In-Reply-To: <20151116183531.90AB4B14091@webabinitio.net> References: <20151116183531.90AB4B14091@webabinitio.net> Message-ID: On Tue, Nov 17, 2015 at 5:35 AM, R. David Murray wrote: >> There are other builders that are stable and should be added to the >> stable builders list. > > Do you have ones you'd like to nominate? AMD64 Debian Root, which I host, has been fairly consistently green. Should that be declared stable? It's proving that the test suite can run with full root privileges. ChrisA From rdmurray at bitdance.com Mon Nov 16 15:46:42 2015 From: rdmurray at bitdance.com (R. David Murray) Date: Mon, 16 Nov 2015 15:46:42 -0500 Subject: [Python-buildbots] Stable builders In-Reply-To: References: <20151116183531.90AB4B14091@webabinitio.net> Message-ID: <20151116204644.202CDB14091@webabinitio.net> On Tue, 17 Nov 2015 07:38:11 +1100, Chris Angelico wrote: > AMD64 Debian Root, which I host, has been fairly consistently green. > Should that be declared stable? It's proving that the test suite can > run with full root privileges. Sound good to me. The ones mentioned in the other email I want to look at more closely as I'm not sure what the status of i390 support is. --David From dje.gcc at gmail.com Mon Nov 16 16:11:54 2015 From: dje.gcc at gmail.com (David Edelsohn) Date: Mon, 16 Nov 2015 16:11:54 -0500 Subject: [Python-buildbots] Stable builders In-Reply-To: <20151116204644.202CDB14091@webabinitio.net> References: <20151116183531.90AB4B14091@webabinitio.net> <20151116204644.202CDB14091@webabinitio.net> Message-ID: On Mon, Nov 16, 2015 at 3:46 PM, R. David Murray wrote: > On Tue, 17 Nov 2015 07:38:11 +1100, Chris Angelico wrote: >> AMD64 Debian Root, which I host, has been fairly consistently green. >> Should that be declared stable? It's proving that the test suite can >> run with full root privileges. > > Sound good to me. > > The ones mentioned in the other email I want to look at more closely as > I'm not sure what the status of i390 support is. David, What does "i390" imply? There is no configuration called i390. The builders are running recent Debian Wheezy, SuSE SLES 12 and Red Hat RHEL 7.1 Linux on z Systems (aka s390x). It's Linux. Let me know if you need more information. Thanks, David From zachary.ware+pydev at gmail.com Mon Nov 16 16:15:56 2015 From: zachary.ware+pydev at gmail.com (Zachary Ware) Date: Mon, 16 Nov 2015 15:15:56 -0600 Subject: [Python-buildbots] Stable builders In-Reply-To: References: Message-ID: On Mon, Nov 16, 2015 at 12:11 PM, David Edelsohn wrote: > What's the plan about updating the list of stable builders? > > The current stable builders don't appear very stable (not all green). > There are other builders that are stable and should be added to the > stable builders list. I had intended to audit all of the builders and come up with a new stable set months ago, but it hasn't happened yet. I'll see whether I can find some time to get anywhere on that this week. -- Zach From rdmurray at bitdance.com Mon Nov 16 17:20:56 2015 From: rdmurray at bitdance.com (R. David Murray) Date: Mon, 16 Nov 2015 17:20:56 -0500 Subject: [Python-buildbots] Stable builders In-Reply-To: References: <20151116183531.90AB4B14091@webabinitio.net> <20151116204644.202CDB14091@webabinitio.net> Message-ID: <20151116222057.3FA60B14095@webabinitio.net> On Mon, 16 Nov 2015 16:11:54 -0500, David Edelsohn wrote: > On Mon, Nov 16, 2015 at 3:46 PM, R. David Murray wrote: > > On Tue, 17 Nov 2015 07:38:11 +1100, Chris Angelico wrote: > >> AMD64 Debian Root, which I host, has been fairly consistently green. > >> Should that be declared stable? It's proving that the test suite can > >> run with full root privileges. > > > > Sound good to me. > > > > The ones mentioned in the other email I want to look at more closely as > > I'm not sure what the status of i390 support is. > > What does "i390" imply? There is no configuration called i390. No, but there is an IBM family called s390 (the i was a typo due to my mixing this up with a customer that uses a different IBM model). > The builders are running recent Debian Wheezy, SuSE SLES 12 and Red > Hat RHEL 7.1 Linux on z Systems (aka s390x). It's Linux. I did not realize that it ran vanilla RHEL/Debian. If there are no modifications to CPython that are specific to the z System, then I would presume there isn't a "supported platform" issue and they could be moved to stable. --David From zachary.ware+pydev at gmail.com Mon Nov 16 18:40:50 2015 From: zachary.ware+pydev at gmail.com (Zachary Ware) Date: Mon, 16 Nov 2015 17:40:50 -0600 Subject: [Python-buildbots] Stable builders In-Reply-To: References: Message-ID: On Mon, Nov 16, 2015 at 3:15 PM, Zachary Ware wrote: > I had intended to audit all of the builders and come up with a new > stable set months ago, but it hasn't happened yet. I'll see whether I > can find some time to get anywhere on that this week. Here's the breakdown as I see it currently: "Stable" (all tests pass on all branches, or have failures that look legitimate): angelico-debian-amd64 bolen-tiger bolen-windows7 edelsohn-debian-z edelsohn-fedora-ppc64 edelsohn-fedora-ppc64le edelsohn-rhel-z edelsohn-sles-z gps-debian-profile-opt gps-ubuntu-exynos5-armv71 kloth-win64 koobs-freebsd9 koobs-freebsd10 langa-ubuntu murray-snowleopard ware-docs ware-gentoo-x86 ware-win81-release "Iffy" (have failures that look related to the slave, but are platforms that should be stable. If I reshuffle right now, they'll be marked unstable): bolen-ubuntu bolen-windows bolen-windows8 bolen-windows10 "Unstable" (have longstanding failures due to slave or platform, or is an explicitly experimental (for python) platform): borja-openbsd-x86 bruynooghe-solaris-csw cea-indiana-x86 edelsohn-aix-ppc64 intel-ubuntu-skylake *-icc koobs-freebsd11 speed-python "Kill" (hasn't been connected in months; I'll remove it): cea-indiana-amd64 Anyone have any thoughts on this before I do the shuffling? Also, I'll note that there's a strange issue with 2.7 that affects three of the otherwise stable bots where the regrtest process doesn't die, leaving buildbot to kill it. -- Zach From db3l.net at gmail.com Mon Nov 16 18:56:32 2015 From: db3l.net at gmail.com (David Bolen) Date: Mon, 16 Nov 2015 18:56:32 -0500 Subject: [Python-buildbots] Stable builders In-Reply-To: References: Message-ID: Just an FYI related to these: > "Iffy" (have failures that look related to the slave, but are > platforms that should be stable. If I reshuffle right now, they'll be > marked unstable): > bolen-ubuntu > bolen-windows > bolen-windows8 > bolen-windows10 I've watched them consistently failing for a while now, but whenever I peek at selected failures it doesn't seem to be slaved related, as opposed to true test failures (albeit those I may not understand a root cause behind). So I've assumed they were issues in-development. Certainly bolen-ubuntu has traditionally been extremely robust in terms of passing tests (unlike windows sometimes) so its failures are usually actual failures. Of course, in that case I suppose a new test might have added host requirements for certificate validation that requires something to change on the build slave. For windows, I sort of assumed the getaddrinfo failures might have been tests that just didn't work properly on newer windows versions, but have to admit I did not try to dig deeper. I don't think there's anything I can do at the slave level to impact that. If the stable list just means passing under the current code, the above set definitely should be excluded. I'm just not sure they aren't representing actual issues for the most part. -- David From rdmurray at bitdance.com Wed Nov 18 11:39:19 2015 From: rdmurray at bitdance.com (R. David Murray) Date: Wed, 18 Nov 2015 11:39:19 -0500 Subject: [Python-buildbots] Stable builders In-Reply-To: References: Message-ID: <20151118163920.EF6DCB14089@webabinitio.net> On Mon, 16 Nov 2015 17:40:50 -0600, Zachary Ware wrote: > On Mon, Nov 16, 2015 at 3:15 PM, Zachary Ware > wrote: > > I had intended to audit all of the builders and come up with a new > > stable set months ago, but it hasn't happened yet. I'll see whether I > > can find some time to get anywhere on that this week. > > Here's the breakdown as I see it currently: > > "Stable" (all tests pass on all branches, or have failures that look > legitimate): > angelico-debian-amd64 > bolen-tiger > bolen-windows7 > edelsohn-debian-z > edelsohn-fedora-ppc64 > edelsohn-fedora-ppc64le > edelsohn-rhel-z > edelsohn-sles-z > gps-debian-profile-opt > gps-ubuntu-exynos5-armv71 > kloth-win64 > koobs-freebsd9 > koobs-freebsd10 > langa-ubuntu > murray-snowleopard > ware-docs > ware-gentoo-x86 > ware-win81-release > > "Iffy" (have failures that look related to the slave, but are > platforms that should be stable. If I reshuffle right now, they'll be > marked unstable): > bolen-ubuntu > bolen-windows > bolen-windows8 > bolen-windows10 > > "Unstable" (have longstanding failures due to slave or platform, or is > an explicitly experimental (for python) platform): > borja-openbsd-x86 > bruynooghe-solaris-csw > cea-indiana-x86 > edelsohn-aix-ppc64 > intel-ubuntu-skylake > *-icc > koobs-freebsd11 > speed-python > > "Kill" (hasn't been connected in months; I'll remove it): > cea-indiana-amd64 > > > Anyone have any thoughts on this before I do the shuffling? Also, > I'll note that there's a strange issue with 2.7 that affects three of > the otherwise stable bots where the regrtest process doesn't die, > leaving buildbot to kill it. I haven't reviewed the logs myself, but looking at what is in the lists based on your judgement, this looks fine to me. I do think we want to resolve the bolen issues and not just move them to unstable, though. It looks like jcea brought the OpenIndiana buildbot back on-line, but it seems to have gotten into trouble processing the backlog. The backlog could just be cancelled; there's not much point in running it. It would be nice to have that bot back, since IIRC it was one of the fastest and so gave quick feedback on bad changesets. However...this is a big enough list of additions to stable that we might want to contact either the release managers or python-dev before making the change. --David From koobs at FreeBSD.org Wed Nov 18 12:04:37 2015 From: koobs at FreeBSD.org (Kubilay Kocak) Date: Thu, 19 Nov 2015 04:04:37 +1100 Subject: [Python-buildbots] Stable builders In-Reply-To: <20151118163920.EF6DCB14089@webabinitio.net> References: <20151118163920.EF6DCB14089@webabinitio.net> Message-ID: <564CAFA5.7060805@FreeBSD.org> On 19/11/2015 3:39 AM, R. David Murray wrote: > It would be nice to have that bot back, since IIRC it was one of the > fastest and so gave quick feedback on bad changesets. If mine are no longer the fastest, I have more work to do :] From rdmurray at bitdance.com Wed Nov 18 13:20:36 2015 From: rdmurray at bitdance.com (R. David Murray) Date: Wed, 18 Nov 2015 13:20:36 -0500 Subject: [Python-buildbots] Stable builders In-Reply-To: <564CAFA5.7060805@FreeBSD.org> References: <20151118163920.EF6DCB14089@webabinitio.net> <564CAFA5.7060805@FreeBSD.org> Message-ID: <20151118182036.8ECA6B14089@webabinitio.net> On Thu, 19 Nov 2015 04:04:37 +1100, Kubilay Kocak wrote: > On 19/11/2015 3:39 AM, R. David Murray wrote: > > It would be nice to have that bot back, since IIRC it was one of the > > fastest and so gave quick feedback on bad changesets. > > If mine are no longer the fastest, I have more work to do :] I did say "one of" :) It's nice to have several that report quickly, because one might just be one of the unstable tests. --David From zachary.ware+pydev at gmail.com Thu Nov 19 01:12:04 2015 From: zachary.ware+pydev at gmail.com (Zachary Ware) Date: Thu, 19 Nov 2015 00:12:04 -0600 Subject: [Python-buildbots] Stable builders In-Reply-To: <20151118163920.EF6DCB14089@webabinitio.net> References: <20151118163920.EF6DCB14089@webabinitio.net> Message-ID: On Wed, Nov 18, 2015 at 10:39 AM, R. David Murray wrote: > I do think we > want to resolve the bolen issues and not just move them to unstable, > though. I agree with that. > It looks like jcea brought the OpenIndiana buildbot back on-line, > but it seems to have gotten into trouble processing the backlog. > The backlog could just be cancelled; there's not much point in > running it. I've cleared the backlog and forced a build on each branch. Since it's running now, I obviously won't be killing it off :). The x86 OpenIndiana bot is also feeling much better now, the 3.4 builder actually completed a successful build. Thank you Jes?s for getting them back to health! I'll consider them for 'stable' after they've had a chance to run a few builds. > It would be nice to have that bot back, since IIRC it was one of the > fastest and so gave quick feedback on bad changesets. > > However...this is a big enough list of additions to stable that we > might want to contact either the release managers or python-dev > before making the change. Here's a slightly revised and annotated listing. I don't think any of the additions should be controversial, though I agree there are a lot of them. I think this is more like how things should be, though. ('+' = added to stable, '-' = removed from stable (none), ' ' = no change) STABLE: +angelico-debian-amd64 - all green +bolen-tiger - all green bolen-ubuntu - all red, possibly a Python bug, possibly network issue bolen-windows - usually green, 2.7 intermittently red bolen-windows7 - all green cea-indiana-amd64 - 2.7, 3.4 green; 3.5+ red (_socket broken) +cea-indiana-x86 - same state as cea-indiana-amd64 +edelsohn-debian-z - 3.any bots green, 2.7 red (regrtest bug) edelsohn-fedora-ppc64 - all green +edelsohn-fedora-ppc64le - all green +edelsohn-rhel-z - all green +edelsohn-sles-z - 3.any bots green, 2.7 red (regrtest bug) +gps-debian-profile-opt - 3.any bots green, 2.7 red (bug) +gps-ubuntu-exynos5-armv71 - all green kloth-win64 - green or legitimate red +koobs-freebsd9 - all green koobs-freebsd10 - all green +langa-ubuntu - all green murray-snowleopard - all green +ware-docs - all green +ware-gentoo-x86 - all green, tests tkinter on Linux +ware-win81-release - mostly green, legitimate bug on 3.4 UNSTABLE: bolen-windows8 - all red (network issue?) bolen-windows10 - all red (network issue?) borja-openbsd-x86 - all red (LibreSSL not officially supported yet) bruynooghe-solaris-csw - non-working hg edelsohn-aix-ppc64 - experimental platform, all red intel-ubuntu-skylake - meant for experimental patch *-icc - still experimental koobs-freebsd11 - not yet started -- Zach From zachary.ware+pydev at gmail.com Thu Nov 19 01:19:28 2015 From: zachary.ware+pydev at gmail.com (Zachary Ware) Date: Thu, 19 Nov 2015 00:19:28 -0600 Subject: [Python-buildbots] Stable builders In-Reply-To: References: Message-ID: On Mon, Nov 16, 2015 at 5:56 PM, David Bolen wrote: > Just an FYI related to these: > >> "Iffy" (have failures that look related to the slave, but are >> platforms that should be stable. If I reshuffle right now, they'll be >> marked unstable): >> bolen-ubuntu >> bolen-windows >> bolen-windows8 >> bolen-windows10 > > I've watched them consistently failing for a while now, but whenever I > peek at selected failures it doesn't seem to be slaved related, as > opposed to true test failures (albeit those I may not understand a > root cause behind). So I've assumed they were issues in-development. > Certainly bolen-ubuntu has traditionally been extremely robust in > terms of passing tests (unlike windows sometimes) so its failures are > usually actual failures. Of course, in that case I suppose a new test > might have added host requirements for certificate validation that > requires something to change on the build slave. I'm not sure about bolen-ubuntu, but I figured its issues are most likely slave- (or slave-environment-) related due to the fact that all branches are failing and the other Ubuntu bots are fine. I did notice that bolen-ubuntu seems to have a pretty old OpenSSL (1.0.1f), might upgrading it be enough? I don't think we support any but the most recent release of any given version of OpenSSL. > For windows, I sort of assumed the getaddrinfo failures might have > been tests that just didn't work properly on newer windows versions, > but have to admit I did not try to dig deeper. I don't think there's > anything I can do at the slave level to impact that. I'm suspicious of Azure for those. My 8.1 bot has no problem with the test that's causing problems on both Azure-based bots. -- Zach (I should have written this message before sending the revised list I just sent a couple minutes ago, I described the state badly for the three bots discussed above :)) From rosuav at gmail.com Thu Nov 19 01:38:17 2015 From: rosuav at gmail.com (Chris Angelico) Date: Thu, 19 Nov 2015 17:38:17 +1100 Subject: [Python-buildbots] Stable builders In-Reply-To: References: <20151118163920.EF6DCB14089@webabinitio.net> Message-ID: On Thu, Nov 19, 2015 at 5:12 PM, Zachary Ware wrote: > +angelico-debian-amd64 - all green Just FYI, I'm currently experiencing a partial network outage, which is impacting my ability to connect to the build farm. So these guys are probably all offline at the moment; they should be up again within a day or so. They should be stable after that. I hate this helpless feeling of "my ISP can't fix this yet". Hopefully it gets sorted out soon... the buildbot isn't the only thing that's down for the moment. ChrisA From db3l.net at gmail.com Thu Nov 19 16:07:42 2015 From: db3l.net at gmail.com (David Bolen) Date: Thu, 19 Nov 2015 16:07:42 -0500 Subject: [Python-buildbots] Stable builders References: Message-ID: Zachary Ware writes: > I'm not sure about bolen-ubuntu, but I figured its issues are most > likely slave- (or slave-environment-) related due to the fact that all > branches are failing and the other Ubuntu bots are fine. I did notice > that bolen-ubuntu seems to have a pretty old OpenSSL (1.0.1f), might > upgrading it be enough? I don't think we support any but the most > recent release of any given version of OpenSSL. Hmm, at the system level it's 1.0.1f, but that's up to date for the platform (Ubuntu 14.04.3) so I'd think should work. I'll do a general update and see if that cleans anything up. >> For windows, I sort of assumed the getaddrinfo failures might have >> been tests that just didn't work properly on newer windows versions, >> but have to admit I did not try to dig deeper. I don't think there's >> anything I can do at the slave level to impact that. > > I'm suspicious of Azure for those. My 8.1 bot has no problem with the > test that's causing problems on both Azure-based bots. Hmm, good point about your bot - I should have thought to look at that. Ok, I'll dig in further. My 8.1 bot had been passing previously (after resolving some of the new build stuff) so figured something new had been introduced, but it's a good point that it might be a collision between the tests and some Azure network restrictions. -- David From zachary.ware+pydev at gmail.com Mon Nov 23 01:58:01 2015 From: zachary.ware+pydev at gmail.com (Zachary Ware) Date: Mon, 23 Nov 2015 00:58:01 -0600 Subject: [Python-buildbots] New installed Python builders Message-ID: Hi, Inspired by a couple of issues about testing installed Python that popped up this morning (#25694 and #25696), I've set up a new set of builders[1] to build, install, and test the installed Python. They run on the same slave that runs my "x86 Gentoo Non-Debug with X" builders. If you are interested, please have a look at how they do what they do, and let me know if you see any issues with how it's set up. Since the 3.x builder is not failing, I'm not sure whether the issues have been fixed or if I messed something up in the setup :) [1] http://buildbot.python.org/all/builders/x86%20Gentoo%20Installed%20with%20X%203.x http://buildbot.python.org/all/builders/x86%20Gentoo%20Installed%20with%20X%203.5 http://buildbot.python.org/all/builders/x86%20Gentoo%20Installed%20with%20X%203.4 http://buildbot.python.org/all/builders/x86%20Gentoo%20Installed%20with%20X%202.7 Regards, -- Zach From rdmurray at bitdance.com Mon Nov 23 11:00:45 2015 From: rdmurray at bitdance.com (R. David Murray) Date: Mon, 23 Nov 2015 11:00:45 -0500 Subject: [Python-buildbots] [python-committers] New installed Python builders In-Reply-To: References: Message-ID: <20151123160046.7D999B140A1@webabinitio.net> On Mon, 23 Nov 2015 00:58:01 -0600, Zachary Ware wrote: > Hi, > > Inspired by a couple of issues about testing installed Python that > popped up this morning (#25694 and #25696), I've set up a new set of > builders[1] to build, install, and test the installed Python. They > run on the same slave that runs my "x86 Gentoo Non-Debug with X" > builders. > > If you are interested, please have a look at how they do what they do, > and let me know if you see any issues with how it's set up. Since the > 3.x builder is not failing, I'm not sure whether the issues have been > fixed or if I messed something up in the setup :) > > [1] http://buildbot.python.org/all/builders/x86%20Gentoo%20Installed%20with%20X%203.x > http://buildbot.python.org/all/builders/x86%20Gentoo%20Installed%20with%20X%203.5 > http://buildbot.python.org/all/builders/x86%20Gentoo%20Installed%20with%20X%203.4 > http://buildbot.python.org/all/builders/x86%20Gentoo%20Installed%20with%20X%202.7 I haven't looked at this, but unless the buildbot does *not* have write access to the installed directories (ie: the install was done as another user) it isn't really doing a full installed python test. --David From rdmurray at bitdance.com Mon Nov 23 16:11:19 2015 From: rdmurray at bitdance.com (R. David Murray) Date: Mon, 23 Nov 2015 16:11:19 -0500 Subject: [Python-buildbots] [python-committers] New installed Python builders In-Reply-To: References: <20151123160046.7D999B140A1@webabinitio.net> Message-ID: <20151123211120.338A1B140A1@webabinitio.net> On Mon, 23 Nov 2015 19:11:15 +0200, Serhiy Storchaka wrote: > On 23.11.15 18:00, R. David Murray wrote: > > On Mon, 23 Nov 2015 00:58:01 -0600, Zachary Ware wrote: > > I haven't looked at this, but unless the buildbot does *not* have write > > access to the installed directories (ie: the install was done as another > > user) it isn't really doing a full installed python test. > > Yes, but at least it catches cases where some files are not installed. > There were few issues with this. True. Something incomplete in this vein is better than nothing. I'm Not sure you should call it "Installed" though, as that will be a bit misleading. Most of the "can't run the tests on installed python" bugs are because the tree is read-only (obviously, not all of them!). Maybe call it "local install"? Wordy, I know, but more accurate. --David From zachary.ware+pydev at gmail.com Fri Nov 27 00:37:40 2015 From: zachary.ware+pydev at gmail.com (Zachary Ware) Date: Thu, 26 Nov 2015 23:37:40 -0600 Subject: [Python-buildbots] [python-committers] New installed Python builders In-Reply-To: <20151123211120.338A1B140A1@webabinitio.net> References: <20151123160046.7D999B140A1@webabinitio.net> <20151123211120.338A1B140A1@webabinitio.net> Message-ID: On Mon, Nov 23, 2015 at 3:11 PM, R. David Murray wrote: > On Mon, 23 Nov 2015 19:11:15 +0200, Serhiy Storchaka wrote: >> On 23.11.15 18:00, R. David Murray wrote: >> > On Mon, 23 Nov 2015 00:58:01 -0600, Zachary Ware wrote: >> > I haven't looked at this, but unless the buildbot does *not* have write >> > access to the installed directories (ie: the install was done as another >> > user) it isn't really doing a full installed python test. >> >> Yes, but at least it catches cases where some files are not installed. >> There were few issues with this. > > True. Something incomplete in this vein is better than nothing. I'm > Not sure you should call it "Installed" though, as that will be a bit > misleading. Most of the "can't run the tests on installed python" bugs > are because the tree is read-only (obviously, not all of them!). > Maybe call it "local install"? Wordy, I know, but more accurate. I've gone with attempting to make it more like a 'real' install, by wrapping the test step with appropriate 'chmod' commands to make the install directory not writable, and confirmed during a test run that the entire installed tree is -w. I also fixed the slave's usage of usePTY (for test_curses) to avoid the failures that had been happening in the 'uninstall' step. -- Zach