Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2017-03-11

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:35 vbellur joined #gluster-dev
01:30 Acinonyx joined #gluster-dev
02:41 riyas joined #gluster-dev
02:48 ilbot3 joined #gluster-dev
02:48 Topic for #gluster-dev is now Gluster Development Channel - http://gluster.org | For general chat go to #gluster | Patches - http://review.gluster.org/ | Channel Logs - https://botbot.me/freenode/gluster-dev/ & http://irclog.perlgeek.de/gluster-dev/
04:29 vimal joined #gluster-dev
04:31 riyas joined #gluster-dev
04:34 mchangir joined #gluster-dev
04:43 Saravanakmr joined #gluster-dev
04:45 msvbhat joined #gluster-dev
04:45 vimal joined #gluster-dev
05:03 msvbhat joined #gluster-dev
05:03 vimal joined #gluster-dev
05:50 rafi joined #gluster-dev
05:54 vimal joined #gluster-dev
05:55 riyas joined #gluster-dev
06:21 riyas joined #gluster-dev
07:05 msvbhat joined #gluster-dev
07:12 susant joined #gluster-dev
08:18 riyas joined #gluster-dev
09:04 msvbhat joined #gluster-dev
09:13 riyas joined #gluster-dev
09:22 jiffin joined #gluster-dev
10:06 msvbhat joined #gluster-dev
10:35 IRCFrEAK joined #gluster-dev
10:35 IRCFrEAK left #gluster-dev
11:33 msvbhat joined #gluster-dev
12:13 jiffin joined #gluster-dev
12:20 rafi joined #gluster-dev
13:10 Saravanakmr joined #gluster-dev
13:42 bfoster joined #gluster-dev
13:53 jiffin joined #gluster-dev
14:05 jiffin joined #gluster-dev
14:20 jiffin1 joined #gluster-dev
14:47 jiffin joined #gluster-dev
15:32 msvbhat joined #gluster-dev
16:00 msvbhat joined #gluster-dev
16:06 msvbhat joined #gluster-dev
16:20 mchangir joined #gluster-dev
16:54 major This can't be right .. neither the git branch release-3.10, nor the git tag v3.10.0 is used for the glusterfs-3.10.0.tar.gz?
17:03 berkayunal joined #gluster-dev
17:04 berkayunal joined #gluster-dev
17:20 major yah .. no .. 'git diff v3.10.0 release-3.10' are distinctly different.. curiously there are changes in the release-3.10 branch that also exist in the Ubuntu deb-src patch series
17:21 major also frustrating is that the source downloads list a glusterfs-3.10.0.tar.gz, but a checksum file for glusterfs-3.10.0rc0 (no checksum for the release tarball)
17:22 misc grmbl, that shouldn't happen
17:23 misc (but we are not yet in a position to automate the release to prevent that :/)
17:24 major I was trying to setup test build so I can test my btrfs changes, but mypatch against release-3.10 seriously fails to apply .. so I have been trying to figure out which branch the tarball was made from so I can pivot my integration branch and have git spit out a new patch..
17:26 major the tarball looks to have been a modified export of some branch .. but it doesn't directly match v3.10.0rc0, v3.10.0rc1, v3.10.0, nor release-3.10
17:31 major hmm.. yah .. the glusterfs-3.10.0-ubuntu1-xenial1 package I am working with has a glusterfs3100.patch which contains some of the delta between v3.10.0 and release-3.10
17:31 major my head esplode
17:32 major and I distincty cannot locate a match which would have produce the released tarball
17:36 misc can you post on glsuter-devel about that ? (cause I am not sure lots of people read the chan on the weekend, and I would prefer to have a archive of that to be refered later...)
17:38 major hmm .. yah .. though atm I am sort of obsessed with just figuring out how to get my build to work cleanly :)
17:38 major half debating abandoning the official tarball at this point
17:38 misc there is maybe a good explication
17:43 major yah .. I dunno .. there is definately something going on .. whole stacks of directories are deleted in the tarball that exist in the repository.. doc/debugging, doc/developer-guide, doc/features .. all gone
17:50 major diff -ruN -x examples -x run-tests-in-vagrant.sh -x tests -x rpc -x rfc.sh -x MAINTAINERS -x .mailmap -x .gitignore -x unittest -x CONTRIBUTING -x geo-replication -x events -x extras -x contrib -x ChangeLog -x build-aux -x doc -x api -x .git -X v3.10.0/.gitignore v3.10.0 glusterfs-3.10.0
17:51 major that reduces the delta down to 3 files .. none of which I am interested in ..
17:51 major I think it shows a pattern of steralizing the "for developers" portion of the source tree
17:51 major though I really can't fathom why..
17:54 ndevos major: release tarballs are made from the git tag, with then configure.ac adjusted for the version, that makes it possible to do versioned builds without the git repository
17:54 ndevos the script that runs in Jenkins is https://github.com/gluster/glusterfs-patc​h-acceptance-tests/blob/master/release.sh
17:55 major yah .. but the tag doesn't match the tarball even a littlebit :(
17:55 ndevos hmm, that sounds weird
17:55 major and .. as mentioned earlier .. there is no checksum released for the tarball .. just one release for a glusterfs-3.10.0rc0 tarball that no longer exists in the downloads
17:56 ndevos the script generates a sha256sum file, that should get copied too... https://github.com/gluster/glusterfs-patch-a​cceptance-tests/blob/master/release.sh#L117
17:57 major https://download.gluster.org/​pub/gluster/glusterfs/LATEST/
17:57 Acinonyx joined #gluster-dev
17:57 ndevos I don't know how that gets synced... nigelb, misc or kkeithley might be able to check that
17:58 ndevos the original gets pushed to http://bits.gluster.org/pub/gluster/glusterfs/src/
17:58 ndevos and the checksum file is there
17:58 major well that answers one mystery :)
17:59 ndevos the console output from the tarball build is at https://build.gluster.org/job/release/187/console
18:00 major cool ..
18:00 ndevos the commit that at that time matched v3.10.0 is d2a3a8f520068dc4e3cb49e2fa49afddb7c5dec7 - and that *should* still be the same
18:01 major as I was saying earlier .. I was setting up pbuilder to build a glusterfs-3.10.0-ubuntu1-xenial1.deb that was patched w/ my btrfs changes so I could test them and my patches against based on the 3.10 branch exploded .. and that lead me down this rabbit hole
18:02 major yah .. part of the problem is I based my branch on release-3.10
18:02 ndevos yeah, I understood that, but I have no idea about the debian builds
18:03 major yah .. they only have 1 patch they are applying .. but part of that patch exists in release-3.10 .. which confused me
18:03 major actually .. all of that patch exists in 3.10
18:03 major erm .. in release-3.10
18:04 major but yah .. the v3.10.0 tag matches the commit from the log
18:04 major I suppose the release.sh script is what is doing the rest of the mangling
18:04 major which explains the rest of the delta
18:05 major time to pivot this integration branch to the tag
18:17 ndevos good luck!
18:24 major just took a rebase and a diff
18:34 major and .. patch applied cleanly and pbuilder is off and running
18:35 major ndevos, thanks for pointing me in the right direction for this stuff
18:37 major now I just need to start putting together test cases on this stuff based on the lvm tests...
18:42 jiffin joined #gluster-dev
18:43 jiffin joined #gluster-dev
18:50 jiffin joined #gluster-dev
19:31 jiffin joined #gluster-dev
19:50 jiffin joined #gluster-dev
19:52 percevalbot joined #gluster-dev
19:52 owlbot joined #gluster-dev
20:04 percevalbot joined #gluster-dev
20:45 owlbot` joined #gluster-dev
20:52 major and damn .. it is a ton of steps to trickle a one-line changes up through 3 topic branches that are based on different parent branches >.<
21:48 major and .. volumes are live and working :)
21:50 major now .. I wager a beer that taking a snapshot is going to explode violently :)
21:52 major okay .. didn't explode .. but failed to snap the volume
21:52 major this damn device name insanity
22:24 major whoa
22:24 major snapshot success
22:25 major just failed post-validation due to a misuse of a variable on my part
22:26 major its the whole mount/unmount into /run/gluster/snaps that failed
22:50 bwerthmann joined #gluster-dev
23:44 bwerthmann joined #gluster-dev

| Channels | #gluster-dev index | Today | | Search | Google Search | Plain-Text | summary