Camelia, the Perl 6 bug

IRC log for #gluster-dev, 2013-03-13

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

All times shown according to UTC.

Time Nick Message
01:02 puebele1 joined #gluster-dev
01:56 hagarth joined #gluster-dev
02:15 sghosh joined #gluster-dev
02:28 kkeithley1 joined #gluster-dev
02:35 jdarcy joined #gluster-dev
02:49 jdarcy joined #gluster-dev
03:06 vshankar joined #gluster-dev
03:21 bharata joined #gluster-dev
03:34 anmol joined #gluster-dev
03:40 sghosh joined #gluster-dev
04:05 bulde joined #gluster-dev
04:18 sahina joined #gluster-dev
04:29 sgowda joined #gluster-dev
04:31 sripathi joined #gluster-dev
04:40 hagarth joined #gluster-dev
04:57 hagarth joined #gluster-dev
05:01 mohankumar joined #gluster-dev
05:14 rastar joined #gluster-dev
05:18 bala joined #gluster-dev
05:27 bharata_ joined #gluster-dev
05:43 lalatenduM joined #gluster-dev
05:51 raghu` joined #gluster-dev
06:00 aravindavk joined #gluster-dev
07:13 aravindavk joined #gluster-dev
07:14 aravindavk joined #gluster-dev
07:16 test_ joined #gluster-dev
07:41 mohankumar joined #gluster-dev
08:43 vshankar_ joined #gluster-dev
08:43 shishir joined #gluster-dev
08:43 test_ joined #gluster-dev
08:43 lala_ joined #gluster-dev
08:44 bulde1 joined #gluster-dev
08:44 kkeithley1 joined #gluster-dev
08:45 bala1 joined #gluster-dev
08:45 rastar1 joined #gluster-dev
08:47 hagarth joined #gluster-dev
08:47 anmol joined #gluster-dev
08:56 rastar joined #gluster-dev
08:56 kkeithley1 joined #gluster-dev
08:56 bulde joined #gluster-dev
08:56 anmol joined #gluster-dev
08:56 lala_ joined #gluster-dev
08:56 hagarth joined #gluster-dev
08:56 spai joined #gluster-dev
08:56 vshankar joined #gluster-dev
08:57 bala joined #gluster-dev
08:57 shishir joined #gluster-dev
09:02 rastar joined #gluster-dev
09:09 sahina joined #gluster-dev
09:09 aravindavk joined #gluster-dev
09:24 puebele1 joined #gluster-dev
09:27 sripathi joined #gluster-dev
09:28 gbrand_ joined #gluster-dev
09:32 hagarth_ joined #gluster-dev
09:34 badone joined #gluster-dev
09:42 jclift joined #gluster-dev
09:43 bharata_ joined #gluster-dev
09:43 maxiz joined #gluster-dev
09:47 sgowda joined #gluster-dev
10:04 bulde1 joined #gluster-dev
10:06 aravindavk joined #gluster-dev
10:27 sripathi joined #gluster-dev
10:29 blues-man joined #gluster-dev
11:10 jdarcy joined #gluster-dev
11:21 jdarcy joined #gluster-dev
11:22 lpabon joined #gluster-dev
11:33 aravindavk joined #gluster-dev
11:44 puebele1 joined #gluster-dev
11:50 sgowda joined #gluster-dev
11:59 edward1 joined #gluster-dev
12:05 JoeJulian joined #gluster-dev
12:17 sripathi joined #gluster-dev
12:20 rastar joined #gluster-dev
12:37 jdarcy joined #gluster-dev
12:38 gbrand_ joined #gluster-dev
12:43 puebele1 joined #gluster-dev
12:54 lalatenduM joined #gluster-dev
12:54 jclift Anyone know how to become a reviewer for gerrit?
12:55 jclift I'm trying to add myself as a review of kkeithley's updated patch set, to +1 it, as it works here.
12:55 jclift But apparently I'm "not a registered user".
13:00 jclift Hmm, it seems that the recent Fedora openid problem is probably what's stuffing me up.
13:13 kkeithley1 joined #gluster-dev
13:30 lalatenduM joined #gluster-dev
13:39 sahina joined #gluster-dev
13:45 sghosh joined #gluster-dev
14:23 wushudoin joined #gluster-dev
14:39 jbrooks joined #gluster-dev
14:41 johnmark jclift: still?
14:41 * johnmark thought that had been resolved
15:05 lpabon joined #gluster-dev
15:18 jclift johnmark: Only partially.
15:19 jclift johnmark: Found a workaround for now though, which is to use the "jc" name when it asks for the reviewer name, instead of using any of the drop down selections.
15:19 jclift johnmark: I've emailed Vijay Bellur about it, as he was trying to help last time. :)
15:55 sahina joined #gluster-dev
16:01 portante joined #gluster-dev
16:26 awheeler_ I asked in the main gluster channel, but this seems more apropos.  Can we add support for '.' in a gluster volume name?  Currently only alpha-numerics , '-', and '_' are valid.
16:26 awheeler_ in cli/src/cli-cmd-parser.c line 214
16:28 awheeler_ And when it finds a non-matching char, it just shows the usage, and doesn't indicate that an invalid char is in the volume name.
16:31 johnmark awheeler_: interesting. I didn't realize that
16:31 johnmark awheeler_: that sounds like a worthy addition. Also doesn't sound like a lot of effort, although I'm wondering if it might break something
16:32 awheeler_ I'm trying to make swauth work, and it ends up wanting volume .auth.
16:32 johnmark awheeler_: would you be able to create a patch?
16:32 awheeler_ Yes.  Just the usual github pull request?
16:33 awheeler_ Or would you rather a patch file?
16:36 ndevos awheeler_: http://www.gluster.org/community/docume​ntation/index.php/Development_Work_Flow
16:37 awheeler_ So, not just a git pull request?  lol
16:37 * ndevos thinks any posix complaint pathname should be acceptible for a volume name
16:38 ndevos awheeler_: nah, but but its a git push :)
16:41 rgustafs joined #gluster-dev
16:44 awheeler_ Ok, I have created the patch, and I'm testing it -- I like this idea better than having swift work-around it.
16:46 hagarth joined #gluster-dev
16:48 jdarcy joined #gluster-dev
16:51 johnmark awheeler_: ah, nice.
17:15 lpabon joined #gluster-dev
17:37 awheeler_ I assume this should be submitted against master?
17:38 jclift awheeler_: If it helps, I've only just recently gone through the above process to get myself set up in Gerrit and so on.
17:39 jclift awheeler_: It's different from the GitHub workflow, and not as intuitive, but it's definitely not hard if following the instructions in that doc.
17:39 awheeler_ It's not hard, just a few more hurdles to clear.  Almost there, (I think)?
17:40 jclift awheeler_: It'll want you sign in with an OpenID provider.  If you don't have a Google account, but you do have a Fedora one, you can use the Fedora OpenID endpoint pretty well.  (ie  http://USERNAME.id.fedoraproject.org)
17:40 jclift awheeler_: Cool, sounds like you're well on the way already then. :D
17:41 awheeler_ I've already got the master repo down, and created a branch so that's all good.
17:41 awheeler_ Trying to understand the tests so I can add one.
17:41 jclift awheeler_: The dev team seems to like things filed as BZ's first, so you have a bug number, and then for the git branch you develop in to use that.  ie bug-123456
17:42 jclift awheeler_: Yeah, I haven't tried out the test stuff yet, so you're ahead of me there.
17:43 * jclift thought it felt a bit weird filing BZ's for (basically) "the build breaks on OSX"... but well, got it done.
17:43 * jclift gets back to work
17:43 awheeler_ Hmm, the BZ part isn't in the doc, lol.
17:44 jclift Yeah, I found out "the hard way" when the review comments were pretty much "this looks good, but please file a BZ and resubmit".
17:44 awheeler_ seemed a bit odd the docs references bug-xyz, but don't tell you where to get xyz.
17:44 jclift So, just saying. ;)
17:44 awheeler_ So, where is the BZ interface?
17:44 jclift Ahhh.  BZ = Bugzilla
17:44 jclift It's at:
17:45 jclift https://bugzilla.redhat.com
17:45 awheeler_ I finally found it, and knew it was bugzilla, just not in the developer docs.  lol
17:46 awheeler_ And will the openid work for BZ?
17:46 jclift I don't think so. :(
17:46 jclift You'll prob need to create an account there first.
17:46 awheeler_ lol
17:46 jclift Once you've got the acct though, to file a Gluster bug it's New -> Community -> GlusterFS.
17:47 awheeler_ This is enough of a barrier I can imagine some people don't submit fixes, or just use github and hope for the best.
17:47 jclift Now that you're mentioning it, yeah, it's not exactly painless is it.
17:48 awheeler_ nope.  I would think that small changes could be submitted via github, and someone would have to walk it through on the other side, and if you start contributing too much, then a full account.  But that's just me.  :)
17:51 jclift It's a decent idea.
17:51 jclift After all, the GitHub repo does have both pull requests and issues enabled.
17:51 jclift They don't seem to get noticed though. :/
17:56 awheeler_ Yes, I see there are several pull requests sitting there.  Didn't look to see how old.
17:56 awheeler_ Bug created.
17:59 jclift Cool.  Sounds like you're in good shape then. :)
18:17 awheeler_ Test case created.
18:18 sghosh joined #gluster-dev
18:20 awheeler_ Hmm, rejected:  ! [remote rejected] HEAD -> refs/for/master/bug-921215 (not Signed-off-by author/committer/uploader)
18:21 awheeler_ I'm guessing that my bug entry has to say that it's ok for this commit to be associated with it?
18:26 mohankumar joined #gluster-dev
18:27 hagarth awheeler_: you just need to git commit --amend -s and resubmit
18:39 awheeler_ ah, ok.
18:40 awheeler_ Hmm, doesn't like that command.
18:41 awheeler_ I'm not familiar with amend nor -s here.  Are there other value should provide?
18:42 awheeler_ Disregard, I can't spell amend apparently.  lol
18:45 jclift :)
18:46 awheeler_ Hmm, when I do the amend, it pops up the usual request for comment.  It's pre-populated, so I just submit as is, and that doesn't seem to be enough, as it's still there if I do it again.
18:47 awheeler_ Do I need to enter something in  addition to the pre-populated text?
18:47 awheeler_ And it's still rejecting my submit
18:48 jclift Hmmm, can you do a git log (showing your patch in it), and fpaste it?
18:48 jclift I just want to see what's showing up in git log for your patch.
18:49 jclift i.e. check if there's a valid "Signed off by" entry and stuff. :)
18:49 awheeler_ http://pastebin.com/Vtc5hXvD
18:49 jclift Note the "Signed off by" string is added by "commit -s".  The amend command above should prob do it too (but I haven't tried).
18:50 jclift Hmmm.  Is the test case supposed to be part of the same patch set as the commit before it?
18:51 * jclift points out that the commit before it doesn't seem to have a "Signed off by" thing.
18:51 awheeler_ Agreed -- how do I get that one amended?
18:52 jclift There's probably an easy way to do it... but I'd generally just cheat and do format-patch on them both to generate standalone patch files, add the "Signed off" string manually, and them re import the patch files.
18:53 jclift An easier approach would be to just git reset (NOT using --hard) back to the commit prior to both of them, and redo the commit as one.
18:53 jclift i.e. git reset caac150b2b0f799a09cab7b3b49498b9dc7ec0e8
18:53 jclift then git add foo bar baz
18:53 jclift git commit -s
18:53 jclift ./rfc.sh
18:54 * jclift uses git reset like this a lot.  There's probably a better way though.
18:56 jclift awheeler_: Reading the comment in your commit, it doesn't sound right either.  This bit sounds half missing "I was modifying the Glusterfs.py code to be able to translate a volume with a name that starts with a '.' to another name, say DOT_volume, but that seemed much messier if we
18:56 jclift Chatted on #gluster-dev and was told this sounded like a good idea by johnmark."
18:58 awheeler_ Cool, that worked.  Thank you very much jclift
18:59 jclift awheeler_: :D
18:59 jclift Now, lets see if it passes review.
19:00 awheeler_ Ah, yes, I see it does have some of the message cut off.
19:00 jclift awheeler_: In my experience, stuff get's reviewed in < 24 hours.
19:00 jclift i.e. I commit stuff, and next day it's reviewed with appropriate feedback/suggestions/approvals/etc.
19:01 * jclift tries to set expectations accordingly :D
19:38 jdarcy joined #gluster-dev
19:48 awheeler_ Ok, cool.
19:49 awheeler_ Well, at least the build was successful.
19:51 awheeler_ But yes, it does look like my comments got cut off.  :(
22:23 awheeler_ It would appear this a harder problem to solve than it looks.  Create is fine, but start is not.
22:56 jdarcy joined #gluster-dev
23:08 awheeler_ johnmark: I have created a patch, but as you suspected, it broke something -- or, more to the point, the volume doesn't start correct, though it doesn't give errors on the command-line to that effect.
23:08 _benoit_ joined #gluster-dev
23:08 awheeler_ johnmark: But now  I can create volumes with a . in them.  lol

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