Closed Bug 847441 Opened 12 years ago Closed 12 years ago

Add remaining win8 machines to the production pool

Categories

(Release Engineering :: General, defect)

x86
Windows 8
defect
Not set
normal

Tracking

(Not tracked)

RESOLVED FIXED

People

(Reporter: armenzg, Assigned: cbook)

References

Details

Attachments

(3 files, 5 obsolete files)

This will require the following changes: * add the slaves to production config [1] * insert them into slavealloc [2] ** make sure to use the values for production rather than preproduction * insert them into graphs [3] ** find the id of "WINNT 6.2 x64" rather than using LAST_INSERT_ID() * reboot the machines into production [4] ** do not reboot them until the first 3 steps are accomplished [1] http://hg.mozilla.org/build/buildbot-configs/file/default/mozilla-tests/production_config.py#l7 [2] https://bugzilla.mozilla.org/attachment.cgi?id=720087 [3] https://bug844130.bugzilla.mozilla.org/attachment.cgi?id=719004 [4] for i in {11..97}; do ipmitool -U <username> -P <password> -H t-w864-ix-0${i}-mgmt.inband.releng.scl3.mozilla.com chassis power soft; done
patch to add the slaves to production config
Attachment #721150 - Flags: review?(armenzg)
Attached patch patch (obsolete) — Splinter Review
updated patch
Attachment #721150 - Attachment is obsolete: true
Attachment #721150 - Flags: review?(armenzg)
Attachment #721169 - Flags: review?(armenzg)
Comment on attachment 721169 [details] [diff] [review] patch Review of attachment 721169 [details] [diff] [review]: ----------------------------------------------------------------- Can you please post one more patch addressing the comments? I will review it right away. ::: mozilla-tests/production_config.py @@ +4,5 @@ > 'xp': dict([("talos-r3-xp-%03i" % x, {}) for x in range(4,10) + range(11,101) \ > if x not in [45, 58, 59]]), # bug 661377, bug 780515, bug 753357 > 'win7': dict([("talos-r3-w7-%03i" % x, {}) for x in range(4,10) + range(11,17) + range(18,105)]), > + 'win8': dict([("t-w864-ix-%03i" % x, {}) for x in range(1,10) + range(11,98) \ > + if x not in [98]]), # IT reserving t-w864-ix-098 to continue to work on the graphics automation An interesting thing about range() in python is that the last number is excluded. e.g. range(1,2) only contains 1. Could you also make it: range(1,98) instead of: range(1,10) + range(11,98) We're not going to leave gaps in our ranges for staging slaves (We used to keep slaves 1 to 3 and 10 as staging slaves). We decided it yesterday in the Monday meeting. If you want increase it to 99 and keep the if below OR keep it at 98 and remove the if statement.
Attachment #721169 - Flags: review?(armenzg) → review-
Attached patch updated patchSplinter Review
review comments adressed
Attachment #721169 - Attachment is obsolete: true
Attachment #721222 - Flags: review?(armenzg)
Comment on attachment 721222 [details] [diff] [review] updated patch Review of attachment 721222 [details] [diff] [review]: ----------------------------------------------------------------- ::: mozilla-tests/production_config.py @@ +4,4 @@ > 'xp': dict([("talos-r3-xp-%03i" % x, {}) for x in range(4,10) + range(11,101) \ > if x not in [45, 58, 59]]), # bug 661377, bug 780515, bug 753357 > 'win7': dict([("talos-r3-w7-%03i" % x, {}) for x in range(4,10) + range(11,17) + range(18,105)]), > + 'win8': dict([("t-w864-ix-%03i" % x, {}) for x in range(1,10) + range(11,99) \ Can you please do "range(1,99)" instead of "range(1,10) + range(11,99)"? r+ with that change.
Attachment #721222 - Flags: review?(armenzg) → review+
(In reply to Armen Zambrano G. [:armenzg] from comment #5) ... > > Can you please do "range(1,99)" instead of "range(1,10) + range(11,99)"? > > r+ with that change. Feel free to land once you make the change.
Attached file csv file for slavealloc addition (obsolete) —
part2: * insert them into slavealloc [2] ** make sure to use the values for production rather than preproduction
Attachment #721649 - Flags: review?(armenzg)
Attachment #721649 - Attachment is patch: true
Attachment #721649 - Attachment mime type: text/csv → text/plain
Attachment #721649 - Attachment is patch: false
Comment on attachment 721649 [details] csv file for slavealloc addition There are two ways to insert data into slavealloc. One of them is as a typical set of mysql insert statement like https://bug844130.bugzilla.mozilla.org/attachment.cgi?id=719005 The other one is through the script "slavealloc dbimport" /tools/slavealloc/bin/slavealloc dbimport -D $db_url --slave-data mydata.csv [1][2] "The CSV file should have the headers specified by 'slavealloc dbimport --help'". I've never tried it myself but here's what the headers should look like: name,distro,bitlength,speed,datacenter,trustlevel,environment,purpose,pool,basedir,enabled talos-r3-xp-096,winxp,32,mini,scl1,try,prod,tests,tests-scl1-windows,C:\\talos-slave,1 You got it similar to the second method but it has to be comma separated and with the header. Otherwise slavealloc dbimport will complain. [1] https://wiki.mozilla.org/ReleaseEngineering/How_To/Create_new_slaves_or_move_them_to_other_pools [2] https://wiki.mozilla.org/ReleaseEngineering:Buildduty:Slave_Management#Adding_a_slave
Attachment #721649 - Flags: review?(armenzg) → review-
Attachment #721649 - Attachment is obsolete: true
Attachment #721707 - Flags: review?(armenzg)
Comment on attachment 721707 [details] new try, this time comma seperated It should be "prod" instead of "pro". It should be "tests" instead of "pool". These values can be determined by visiting http://slavealloc.build.mozilla.org/ui/#slaves and filtering by "t-w864". Please remove t-w864-ix-042 (it already exists) and t-w864-ix-098 (we won't be using it) from that list. r+ with those changes. Please use "slavealloc dbimport" method when importing the data. Let me know if you would like to share a screen session on slavealloc so I can guide you with it.
Attachment #721707 - Flags: review?(armenzg) → review+
Hey Armen, thanks for the review, so the docs tell for db url [root@slavealloc ~]# /tools/slavealloc/bin/slavealloc dbimport -D $db_url --slave-data win8slavealloc2.csv usage: slavealloc dbimport [-h] [-D DBURL] [--slave-data SLAVE_DATA] [--master-data MASTER_DATA] [--master-json MASTER_JSON] [--password-data PASSWORD_DATA] [--enable] slavealloc dbimport: error: argument -D/--db: expected one argument here $db_url is most easily found in slavealloc's shell history, do -D $db is -d devtools-rw-vip.db.scl3.mozilla.com right ?
(In reply to Carsten Book [:Tomcat] from comment #12) > Hey Armen, thanks for the review, so the docs tell for db url > > [root@slavealloc ~]# /tools/slavealloc/bin/slavealloc dbimport -D $db_url > --slave-data win8slavealloc2.csv > usage: slavealloc dbimport [-h] [-D DBURL] [--slave-data SLAVE_DATA] > [--master-data MASTER_DATA] > [--master-json MASTER_JSON] > [--password-data PASSWORD_DATA] [--enable] > slavealloc dbimport: error: argument -D/--db: expected one argument > > here $db_url is most easily found in slavealloc's shell history, do -D $db > is -d devtools-rw-vip.db.scl3.mozilla.com right ? and done!
Attachment #721707 - Flags: checked-in+
Attached file graph server patch/csv file (obsolete) —
Attachment #722233 - Flags: review?(armenzg)
Attached file graph server patch/csv file (obsolete) —
Attachment #722233 - Attachment is obsolete: true
Attachment #722233 - Flags: review?(armenzg)
Attachment #722235 - Flags: review?(armenzg)
Comment on attachment 722235 [details] graph server patch/csv file Those have to be mysql insert statements.
Attachment #722235 - Flags: review?(armenzg) → review-
Attachment #722235 - Attachment is obsolete: true
Attachment #722666 - Flags: review?(armenzg)
Attachment #722666 - Flags: review?(armenzg) → review+
Attachment #722666 - Flags: checked-in+
Once we have a reconfig we should be able to add the machines into production by running this: [4] for i in {11..97}; do ipmitool -U <username> -P <password> -H t-w864-ix-0${i}-mgmt.inband.releng.scl3.mozilla.com chassis power soft; done
Whiteboard: waiting on reconfig to reboot machines into production
Merged and reconfiguration completed.
Whiteboard: waiting on reconfig to reboot machines into production → TODO: reboot machines into production
(In reply to Armen Zambrano G. [:armenzg] from comment #18) > Once we have a reconfig we should be able to add the machines into > production by running this: > [4] for i in {11..97}; do ipmitool -U <username> -P <password> -H > t-w864-ix-0${i}-mgmt.inband.releng.scl3.mozilla.com chassis power soft; done and done! its just doing the power soft!
Whiteboard: TODO: reboot machines into production
It seems that we had to run "chassis power reset" instead. I have taken care of this and all win8 machines are now in production.
Status: NEW → RESOLVED
Closed: 12 years ago
Resolution: --- → FIXED
Product: mozilla.org → Release Engineering
Component: General Automation → General
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: