- OpenStack-I18n - lists.openstack.org

Re: [OpenStack-I18n] [OpenStack-Infra] translate-dev wildfly/zanata/trove issues
by Ian Y. Choi 15 Jan '17

15 Jan '17

Thanks a lot for such sharing Ian! - not me :) I am copying to i18n mailing list and Zanata dev team for FYI purpose (as what Kato-san kindly RST markups for option on another mail thread). Note that this issue is re-generatable: I am able recreate the issue on translate-dev : When I create a new version from openstack-manuals from master branch [1] - 20107950 words, there is no further web responses from translate-dev.o.o around after 3/4 of total words were processed like [2-4]. It was completely fine for a version creation with 2610150 words. If someone has a good idea to further investigate on it, please share to us :) [1] https://translate-dev.openstack.org/iteration/view/openstack-manuals/master [2] https://translate-dev.openstack.org/iteration/view/openstack-manuals/stable… [3] https://translate-dev.openstack.org/iteration/view/openstack-manuals/test-v… [4] https://translate-dev.openstack.org/iteration/view/openstack-manuals/test-v… With many thanks, /Ian Ian Wienand wrote on 1/16/2017 12:03 PM: > Hi, > > I was alerted to translate-dev performance issues today. Indeed, it > seemed that things were going crazy with the java wildfly process > sucking up all CPU. > > At first there didn't seem to be anything in the logs. Java was > clearly going mad however, with the following threads going flat-out. > > --- gone crazy processes --- > 14807 wildfly 20 0 9.892g 4.729g 34232 R 93.7 60.7 26:56.73 java > 14804 wildfly 20 0 9.892g 4.729g 34232 R 92.4 60.7 26:51.91 java > 14806 wildfly 20 0 9.892g 4.729g 34232 R 92.4 60.7 26:53.92 java > 14808 wildfly 20 0 9.892g 4.729g 34232 R 92.4 60.7 26:53.97 java > 14810 wildfly 20 0 9.892g 4.729g 34232 R 92.4 60.7 26:56.28 java > 14809 wildfly 20 0 9.892g 4.729g 34232 R 92.1 60.7 26:57.74 java > 14803 wildfly 20 0 9.892g 4.729g 34232 R 91.1 60.7 26:54.90 java > 14805 wildfly 20 0 9.892g 4.729g 34232 R 90.4 60.7 26:52.44 java > --- > > Hoping to find a easy smoking-gun, I made the java process dump it's > threads to see what these are doing > > --- thread dump --- > > 14807 14804 14806 14808 14810 14809 14803, 14805 > 0x39d7 0x39d4 0x39d6 0x39d8 0x39da 0x39d9 0x39d3 0x39d5 > > "GC task thread#4 (ParallelGC)" os_prio=0 tid=0x00007fb2a8026000 nid=0x39d7 runnable > "GC task thread#4 (ParallelGC)" os_prio=0 tid=0x00007fb2a8026000 nid=0x39d7 runnable > "GC task thread#4 (ParallelGC)" os_prio=0 tid=0x00007fb2a8026000 nid=0x39d7 runnable > "GC task thread#4 (ParallelGC)" os_prio=0 tid=0x00007fb2a8026000 nid=0x39d7 runnable > > "GC task thread#1 (ParallelGC)" os_prio=0 tid=0x00007fb2a8021000 nid=0x39d4 runnable > "GC task thread#1 (ParallelGC)" os_prio=0 tid=0x00007fb2a8021000 nid=0x39d4 runnable > "GC task thread#1 (ParallelGC)" os_prio=0 tid=0x00007fb2a8021000 nid=0x39d4 runnable > "GC task thread#1 (ParallelGC)" os_prio=0 tid=0x00007fb2a8021000 nid=0x39d4 runnable > > "GC task thread#3 (ParallelGC)" os_prio=0 tid=0x00007fb2a8024800 nid=0x39d6 runnable > "GC task thread#3 (ParallelGC)" os_prio=0 tid=0x00007fb2a8024800 nid=0x39d6 runnable > "GC task thread#3 (ParallelGC)" os_prio=0 tid=0x00007fb2a8024800 nid=0x39d6 runnable > "GC task thread#3 (ParallelGC)" os_prio=0 tid=0x00007fb2a8024800 nid=0x39d6 runnable > > "GC task thread#5 (ParallelGC)" os_prio=0 tid=0x00007fb2a8028000 nid=0x39d8 runnable > "GC task thread#5 (ParallelGC)" os_prio=0 tid=0x00007fb2a8028000 nid=0x39d8 runnable > "GC task thread#5 (ParallelGC)" os_prio=0 tid=0x00007fb2a8028000 nid=0x39d8 runnable > "GC task thread#5 (ParallelGC)" os_prio=0 tid=0x00007fb2a8028000 nid=0x39d8 runnable > > "GC task thread#7 (ParallelGC)" os_prio=0 tid=0x00007fb2a802b800 nid=0x39da runnable > "GC task thread#7 (ParallelGC)" os_prio=0 tid=0x00007fb2a802b800 nid=0x39da runnable > "GC task thread#7 (ParallelGC)" os_prio=0 tid=0x00007fb2a802b800 nid=0x39da runnable > "GC task thread#7 (ParallelGC)" os_prio=0 tid=0x00007fb2a802b800 nid=0x39da runnable > > "GC task thread#6 (ParallelGC)" os_prio=0 tid=0x00007fb2a8029800 nid=0x39d9 runnable > "GC task thread#6 (ParallelGC)" os_prio=0 tid=0x00007fb2a8029800 nid=0x39d9 runnable > "GC task thread#6 (ParallelGC)" os_prio=0 tid=0x00007fb2a8029800 nid=0x39d9 runnable > "GC task thread#6 (ParallelGC)" os_prio=0 tid=0x00007fb2a8029800 nid=0x39d9 runnable > > "GC task thread#0 (ParallelGC)" os_prio=0 tid=0x00007fb2a801f000 nid=0x39d3 runnable > "GC task thread#0 (ParallelGC)" os_prio=0 tid=0x00007fb2a801f000 nid=0x39d3 runnable > "GC task thread#0 (ParallelGC)" os_prio=0 tid=0x00007fb2a801f000 nid=0x39d3 runnable > "GC task thread#0 (ParallelGC)" os_prio=0 tid=0x00007fb2a801f000 nid=0x39d3 runnable > --- > > Unfortunately, these all appear to be GC threads so there's nothing > really obvious there. > > However, eventually in the log you start getting stuff like > > --- > 2017-01-16T01:58:00,844Z SEVERE [javax.enterprise.resource.webcontainer.jsf.application] (default task-43) Error Rendering View[/project/project.xhtml]: javax.el.ELException: /WEB-INF/layout/project/settings-tab-languages.xhtml @117,88 rendered="#{not projectHome.hasLocaleAlias(locale)}": javax.persistence.PersistenceException: org.hibernate.HibernateException: Transaction was rolled back in a different thread! > ... > 2017-01-16T01:08:48,805Z WARN [org.hibernate.engine.jdbc.spi.SqlExceptionHelper] (pool-5-thread-1) SQL Error: 0, SQLState: null > 2017-01-16T01:08:48,806Z ERROR [org.hibernate.engine.jdbc.spi.SqlExceptionHelper] (pool-5-thread-1) javax.resource.ResourceException: IJ000460: Error checking for a transaction > 2017-01-16T01:08:48,813Z ERROR [org.hibernate.AssertionFailure] (pool-5-thread-1) HHH000099: an assertion failure occured (this may indicate a bug in Hibernate, but is more likely due to unsafe use of the session): org.hibernate.exception.GenericJDBCException: Could not open connection > 2017-01-16T01:08:48,813Z WARN [com.arjuna.ats.jta] (pool-5-thread-1) ARJUNA016029: SynchronizationImple.afterCompletion - failed for org.hibernate.engine.transaction.synchronization.internal.RegisteredSynchronization@3fc123aa with exception: org.hibernate.AssertionFailure: Exception releasing cache locks > --- > > There's literally hundreds of lines of traceback across various > exceptions, but at the bottom this really seems to point to mysql > being at the root of all this? > > In this case, mysql is running in RAX's cloud db service. I logged > into the RAX control-panel and everything for the DB was green. I > used the same credentials as zanata to authenticate and I can connect > to the db via translate-dev.o.o and do a few basic things. > > --- > MySQL [(none)]> USE zanata; > Reading table information for completion of table and column names > You can turn off this feature to get a quicker startup with -A > > Database changed > MySQL [zanata]> show tables; > +----------------------------------+ > | Tables_in_zanata | > +----------------------------------+ > | Activity | > | DATABASECHANGELOG | > ... > --- > > So it's not as simple as "db not working" :( I have copied the console > log to [1] in case that makes any sense to zanata people. > > If this persists, and isn't a known Zanata issue, I think we might > have to enable root access for the db via the API which might allow us > to do some thing like debug long-queries. Otherwise, there might be > something in the mysql logs available to RAX support people that would > otherwise help identify the issue. OTOH, maybe it's as simple as > upping java heaps or something. > > In the mean time, I have restarted wildfly which seems to get things > going for a while. > > -i > > [1] http://people.redhat.com/~iwienand/console.log.translate-dev-2017-01-16.log… > > _______________________________________________ > OpenStack-Infra mailing list > OpenStack-Infra(a)lists.openstack.org > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-infra

2 1

FYI: RST markups for option
by Kato, Tomoyuki 15 Jan '17

15 Jan '17

Hi I18n team, Docs team changed convention about RST markup for options[1]. The newer version of Sphinx warns about :option:` ` usage without .. option:: description. Since we do not use .. option:: to describe parameter, remove :option:` `. Please keep in mind using :option:` ` breaks build in traslating. [1] http://docs.openstack.org/contributor-guide/rst-conv/inline-markups.html#op… Regards, KATO Tomoyuki

1 0

Re: [OpenStack-I18n] [openstack-dev] [release][ptl] not planning to serve as release PTL for Pike
by Ian Y. Choi 14 Jan '17

14 Jan '17

Thank a lot for your great work on release management team. Also, I would like to express my appreciation to Doug especially for communications with I18n team. I remember that Doug kindly came to I18n room at Austin Design Summit, and we discussed some release management issues for I18n team. Moreover, releasenotes translation support would not be possible without your kind assistance. With many thanks, Doug Hellmann wrote on 1/14/2017 5:00 AM: > I announced this at the release team meeting on 6 Jan, but thought > I should also post to the list as well: I do not plan to serve as > the Release Management team PTL for the Pike release cycle. > > It has been my pleasure to serve as PTL, and we've almost finished > the automation work that I envisioned when I joined the team. Now > I would like to shift my focus to some other projects within the > community. I will still be an active member of the Release Management > team, helping with new initiatives and reviews for releases. > > Doug > > __________________________________________________________________________ > OpenStack Development Mailing List (not for usage questions) > Unsubscribe: OpenStack-dev-request@lists.openstack.org?subject:unsubscribe > http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev

1 0

PTG planning etherpad
by Alexandra Settle 13 Jan '17

13 Jan '17

Hi everyone, I have created an etherpad for i18n and docs to collaborate on: https://etherpad.openstack.org/p/docs-i18n-ptg-pike I have added this etherpad to the wiki page: https://wiki.openstack.org/wiki/PTG/Pike/Etherpads#Monday_-.3E_Tuesday Please allocate if you are able to attend in the ‘attendees’ section. If you have any topics you wish to discuss during the PTG (for docs or i18n) please list them in the etherpad ☺ Thanks, Alex

3 2

What's Up, Doc? 13 January
by Lana Brindley 12 Jan '17

12 Jan '17

Hi everyone, Welcome to 2017. I hope you all had a chance to rest up a little before we hurtle straight into another release, as Ocata is now only 40 days away! I'll be in contact with our lovely release managers Brian and Mariia over the next week or so to kick off release management tasks. We will need people to test the Installation Guide again, too, so if you're interested in helping out there, please let me know. As many of you already know, I'm not planning on running for Pike PTL. PTL nominations open in the week of 23 January, and I'll be working with Alex and Kato-san who have both expressed interest in taking on this role. If you're interested in running for PTL (for Pike or any future release), and would like a crash course in PTLing, please let me know and I can share whatever meagre wisdom I've gleaned. == Progress towards Ocata == 40 days to go! Closed 149 bugs so far. Deliverables are being tracked here: https://wiki.openstack.org/wiki/Documentation/OcataDeliverables == The Road to PTG in Atlanta == Event info is available here: http://www.openstack.org/ptg Purchase tickets here: https://pikeptg.eventbrite.com/ Docs is a horizontal project, so our sessions will run across the Monday and Tuesday of the event. We will be combining the docs event with i18n, so translators and docs people will all be in the room together. The incoming PTL will be responsible for planning and running sessions at the Atlanta PTG. Also, a quick note that the CFP and ticket sales for *Boston* in May are now open: https://www.openstack.org/summit/boston-2017/call-for-presentations/ == Speciality Team Reports == * API: Anne Gentle Fairy-slipper is being retired. It was a good tool, served its purpose, and can now traipse into the sunset, thanks to AJaeger and stevemar for the help with the patches. Be sure to follow the service catalog naming conventions work going on in the API WG. And there's also a new EC2 API reference: http://developer.openstack.org/api-ref/ec2-api/ * Configuration Reference and CLI Reference: Tomoyuki Kato Config Reference: Updated a few plug-ins. CLI reference: Updated a few command references and removed unsupported Block Storage API v1 and Image API v1. * High Availability Guide: Andrew Beekhof No report this week * Hypervisor Tuning Guide: Joe Topjian No report this week * Installation guides: Lana Brindley Testing for Ocata will be opening soon. Also, we need a web dev for the landing page, if you know of anyone, please get me in touch. * Networking Guide: John Davidge No update this week. * Operations and Architecture Design guides: Darren Chan Ben Silverman has spent some time refining the Compute content in the Arch Guide draft. Thanks Ben! We are also looking for more SMEs to help with the Arch Guide. So if you're interested, contact asettle or me. * Security Guide: Nathaniel Dillon No update this week. * Training Guides: Matjaz Pancur No update this week. * Training labs: Pranav Salunke, Roger Luethi No update this week. * User guides: Joseph Robinson The command conversions have happened at a good pace, the User Guide appears to be mostly complete, with the condition that the trove and manila teams do not have OpenStack commands yet. I have started checking through the Admin Guide files. I suspect there is more to convert, but it should be ready for Ocata. == Doc team meeting == Our next meeting will be on Thursday 27 January at 2100 UTC in #openstack-meeting-alt For more meeting details, including minutes and the agenda: https://wiki.openstack.org/wiki/Meetings/DocTeamMeeting -- Keep on doc'ing! Lana https://wiki.openstack.org/wiki/Documentation/WhatsUpDoc#13_January_2017 -- Lana Brindley Technical Writer Rackspace Cloud Builders Australia http://lanabrindley.com

1 0

Discussion: How to test Zanata well for translate-dev.openstack.org?
by Ian Y. Choi 12 Jan '17

12 Jan '17

Hello, First of all, I am really sorry for raising this issue late. (I have thought that I already shared it through openstack-i18n mailing list, but it was actually in my draft mailbox..) Previously, when I18n team migrated translation platform from Transifex to Zanata, http://translate-dev.openstack.org was used to test Zanata. Current Zanata version we use is 3.7.3, and now I18n wants to have newer Zanata version (e.g., 3.9.6: https://github.com/zanata/zanata-platform/releases/tag/server-3.9.6 - current latest) with a few bug fixes and newer functionalities. Infra team is now helping to upgrade Zanata with server operating system upgrade (Current translate.openstack.org uses Trusty - Ubuntu 14.04 but newer Zanata version needs Xenial - Ubuntu 16.04), and now http://translate-dev.openstack.org is running with current Zanata 3.7.3 version and *Xenial*! For the first step, I18n team needs to test translate-dev.openstack.org - whether Zanata 3.7.3 works with Xenial operating system or not. Please see possible test items on Etherpad: https://etherpad.openstack.org/p/i18n-zanata-test-for-translate-dev Actually, since I did not involve too much on previously migration from Transifex to Zanata, I am not sure whether current test items on Etherpad is sufficient for Zanata test or not. So, please review test items for such testing purposes and add or modify items if needed. Also, involving translate-dev test activities and updating test result with 'O' mark in Etherpad will be really appreciated. If our test result is fine, then the next step would be to test Zanata latest version with Xenial on translate-dev.openstack.org . With many thanks, /Ian

5 6

Re: [OpenStack-I18n] [I18n] Sharing Documentation translation stats
by C. Verheyde 11 Jan '17

11 Jan '17

Hi, Thank you for the sharing of the stats Ian ! That's great to see it progressing. I noticed a "bug" in identity in stackalytics. My nickname/logname is Corin, but it seems I'm refered in stackalytics as "Corin Royal Drummond<http://stackalytics.com/?metric=translations&release=all&user_id=corin>". I don't know at all where this "Royal Drummond" suffix comes from ??? http://stackalytics.com/?metric=translations&release=ocata) Regards, Corinne

3 4

[I18n] Sharing Documentation translation stats during Ocata cycle
by Ian Y. Choi 11 Jan '17

11 Jan '17

Hello! When I went to last Barcelona Summit, I told that I would share updates from I18n to Docs mailing list monthly. But, in reality, my daily and usual stuffs prevented me from writing a good and nice article as Lana does so well. Furthermore, time is really going too fast (Half of Ocata cycle is already passed!). Now I would like to share Documentation translation statistics from translators for last couple of months in Ocata cycle to Docs and I18n team. Note that my sharing might not be perfect. I have not quantitatively measured how many words are in OpenStack documents and also how translation qualities are improved by review activities, but I welcome your feedback from my first sharing. I would like to keep sharing such updates with Docs team :) == Documentation translation plan during Ocata cycle == - By the last December 2016, I18n team set higher priority for the following document translation: : (stable-newton) openstack-manuals: install-guide : (stable-newton) openstack-manuals: common : (stable-newton) openstack-manuals: networking-guide : (master) openstack-manuals: user-guide : (master) training-guides: upstream-training == Statistics data for documentation translations == - By now, there were total 158,414 translated words in openstack-manuals for Ocata cycle. Many translators from different countries have participated in such translation. I really appreciate such kind translation contribution! - Also, there were 23,552 translated words in training-guides (upstream-training) and 1,159 words for api-site (api-quick-start) for OpenStack documents. - For more information on statistics, please see translation metrics in Stackalytics (URL: http://stackalytics.com/?metric=translations&release=ocata) == Completed documentation translation == The following documents have significant progress during Ocata cycle. I have checked the percentage of translations by documents on late November 2016 and just today from Zanata, OpenStack translation platform: https://translate.openstack.org/ . - openstack-manuals in stable/newton : common => 100% translated for Indonesian, Korean, and Russian! : install-guide => Now French, Indonesian, and Korean documents are fully translated. => Also, there have been lots of translations contributed from German translators. - openstack-manuals in master : Japanese translators have constantly contributed their translations. Now the following documents are fully translated into Japanese => ha-guide, image-guide, and user-guide : Also, Indonesian user-guide is 100% translated. - training-guides : Indonesian upstream-training is now fully translated. : Nice progress for German (45.04% -> 74.33%) More detail translation progress with published pages is available at : https://etherpad.openstack.org/p/i18n-translation-status-documentation . == Published translated documents and Language landing pages == It is important for language landing pages to have up-to-date links for published translated documents. During Ocata cycle, - French landing page now includes Mitaka install-guide (Note: It was fully translated before Ocata cycle, but translation reviews by French translators have been accomplished during Ocata cycle). - Korean landing page now includes Newton install-guide. For language landing pages, the following modifications will be accomplished soon hopefully. Note that it will depend on specific translation and review plan from different languages. : Indonesian landing page will include more published translated documents. : Japanese landing page will include ha-guide, which is now 100% translated. : French landing page will be refined with a new theme (the one now docs.openstack.org uses). That's what I wanted to share with Docs and I18n team for documentation translation statistics. Now Ocata Dashboard translation including Horizon is getting more important for I18n team, since StringFreeze periods are relatively short in Ocata cycle. However, that does not mean that documentation translation for I18n team has been finished in Ocata cycle. Although translators would be busy for translation activities, I would like to encourage to also spend time on translation reviews to make translated documents with higher qualities. With many thanks, /Ian

3 2

Re: [OpenStack-I18n] Puzzled with Japanese ha-guide
by Kato, Tomoyuki 06 Jan '17

06 Jan '17

Hi Andreas, Good finding. I submitted a patch: https://review.openstack.org/#/c/417336/ Regards, KATO Tomoyuki > https://review.openstack.org/#/c/417318/ imports the Japanese ha-guide > and trying to build it, sphinx fails with: > > Exception occurred: > File > "/home/aj/Software/vcs/OpenStack/openstack/openstack-manuals/.tox/venv/lib/python2.7/site-packages/docutils/nodes.py", > line 1367, in set_duplicate_name_id > old_node = self.ids[old_id] > KeyError: 'install-rabbitmq' > > > I'm not sure what's going on here ;( > > we have in shared-messaging-rst: > - :ref:`Install RabbitMQ<rabbitmq-install>` > > And later the headline "Install RabbitMQ" with the label above. > > If I remove this entry from the po file, it still fails building: > > msgid ":ref:`Install RabbitMQ<rabbitmq-install>`" > msgstr ":ref:`RabbitMQ のインストール<rabbitmq-install>`" > > Could somebody investigate and fix this, please? [snip]

2 1

Puzzled with Japanese ha-guide
by Andreas Jaeger 06 Jan '17

06 Jan '17

https://review.openstack.org/#/c/417318/ imports the Japanese ha-guide and trying to build it, sphinx fails with: Exception occurred: File "/home/aj/Software/vcs/OpenStack/openstack/openstack-manuals/.tox/venv/lib/python2.7/site-packages/docutils/nodes.py", line 1367, in set_duplicate_name_id old_node = self.ids[old_id] KeyError: 'install-rabbitmq' I'm not sure what's going on here ;( we have in shared-messaging-rst: - :ref:`Install RabbitMQ<rabbitmq-install>` And later the headline "Install RabbitMQ" with the label above. If I remove this entry from the po file, it still fails building: msgid ":ref:`Install RabbitMQ<rabbitmq-install>`" msgstr ":ref:`RabbitMQ のインストール<rabbitmq-install>`" Could somebody investigate and fix this, please? Note for building, I did: diff --git a/doc-tools-check-languages.conf b/doc-tools-check-languages.conf index d562f83..ddb0047 100644 --- a/doc-tools-check-languages.conf +++ b/doc-tools-check-languages.conf @@ -8,7 +8,8 @@ declare -A BOOKS=( ["de"]="install-guide" ["fr"]="install-guide" ["id"]="install-guide user-guide" - ["ja"]="image-guide install-guide networking-guide ops-guide user-guide" +# ["ja"]="ha-guide image-guide install-guide networking-guide ops-guide user-guide" + ["ja"]="ha-guide" ["ko_KR"]="install-guide" ["zh_CN"]="install-guide user-guide" ) @@ -19,7 +20,7 @@ declare -A DRAFTS=( ["de"]="install-guide" ["fr"]="install-guide" ["id"]="install-guide" - ["ja"]="install-guide networking-guide ops-guide" + ["ja"]="ha-guide install-guide networking-guide ops-guide" ["ko_KR"]="install-guide" ["zh_CN"]="install-guide" And then run locally "tox -e buildlang -- ja". Note that it gives a success message despites failing (the next openstack-doc-tools update will fail this), Andreas -- Andreas Jaeger aj(a){suse.com,opensuse.org} Twitter: jaegerandi SUSE LINUX GmbH, Maxfeldstr. 5, 90409 Nürnberg, Germany GF: Felix Imendörffer, Jane Smithard, Graham Norton, HRB 21284 (AG Nürnberg) GPG fingerprint = 93A3 365E CE47 B889 DF7F FED1 389A 563C C272 A126

1 0