Camelia, the Perl 6 bug

IRC log for #gluster-dev, 2013-09-19

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

All times shown according to UTC.

Time Nick Message
00:30 awheeler joined #gluster-dev
00:31 awheeler joined #gluster-dev
01:03 bala joined #gluster-dev
01:06 lalatenduM joined #gluster-dev
01:11 fandikurnia01 joined #gluster-dev
02:17 bharata-rao joined #gluster-dev
02:53 mohankumar joined #gluster-dev
03:05 kshlm joined #gluster-dev
03:06 fandikurnia01 joined #gluster-dev
03:32 shubhendu joined #gluster-dev
03:47 itisravi joined #gluster-dev
04:01 vshankar joined #gluster-dev
04:08 mohankumar joined #gluster-dev
04:15 ppai joined #gluster-dev
04:17 ndarshan joined #gluster-dev
04:23 hagarth joined #gluster-dev
04:26 kanagaraj joined #gluster-dev
04:40 mohankumar joined #gluster-dev
04:43 ndarshan joined #gluster-dev
04:51 bala joined #gluster-dev
04:52 aravindavk joined #gluster-dev
05:00 lalatenduM joined #gluster-dev
05:14 kanagaraj joined #gluster-dev
05:19 ajha joined #gluster-dev
05:46 bala joined #gluster-dev
05:51 mohankumar joined #gluster-dev
06:01 mohankumar joined #gluster-dev
06:02 ndarshan joined #gluster-dev
06:03 ppai joined #gluster-dev
06:11 itisravi joined #gluster-dev
06:12 kshlm joined #gluster-dev
06:12 sac joined #gluster-dev
06:12 sac`away joined #gluster-dev
06:12 aravindavk joined #gluster-dev
06:12 shubhendu joined #gluster-dev
06:12 ajha joined #gluster-dev
06:12 vshankar joined #gluster-dev
06:13 hagarth joined #gluster-dev
06:19 lalatenduM joined #gluster-dev
06:20 bulde joined #gluster-dev
06:22 kanagaraj joined #gluster-dev
06:28 ppai joined #gluster-dev
06:28 ndarshan joined #gluster-dev
06:32 mohankumar joined #gluster-dev
06:34 bala joined #gluster-dev
06:46 ndarshan joined #gluster-dev
07:29 raghu joined #gluster-dev
07:50 kshlm joined #gluster-dev
07:50 kshlm joined #gluster-dev
08:40 hagarth joined #gluster-dev
09:24 ppai joined #gluster-dev
09:36 bulde1 joined #gluster-dev
09:43 ppai joined #gluster-dev
10:35 vshankar joined #gluster-dev
10:52 shubhendu joined #gluster-dev
11:17 jclift joined #gluster-dev
11:40 jclift ndevos: Sorry I haven't been able to comment on the regression-tests package yet.  Was busy yesterday, and now (in office with new F19 laptop) it turns out I've forgotten to copy my ssh keys and similar to it.
11:40 jclift ndevos: I think the package itself is a good idea.  But I think it *should* be included in the main Fedora packages, so we have consistency.
11:42 jclift ndevos: Probably what we should do is have the run-tests.sh script display a big warning message at the start about "**** This will delete all GlusterFS data on this server ****" plus require explicit confirmation to continue.
11:42 jclift ndevos: eg make it very obvious what will happen.
11:43 shubhendu joined #gluster-dev
11:43 jclift ndevos: .... hmmm, we should probably include a "press ^C to exit" option or similar too at that warning point, as well as let the script detect scripted mode too
11:43 jclift eg.  run-tests.sh  --mode=scripted (or whatever the standard Gluster option is for that, I don't remember atm)
11:44 jclift kkeithley kkeithley_: ^^^ How does the above sound to you?  Good/bad/other? :)
11:47 ndevos jclift: yes, that would be needed, and somehow we would need to compile all the .c (and other?) files that come withe the test-suite, /usr/share should be see and read-only during runtime
11:48 ndevos jclift: and about you're touchpad, you can just unbind it from the command-line (reboot if your keyboard stops working)
11:49 ppai joined #gluster-dev
11:49 ndevos jclift: in case you're in the bakerstreet office, I'm sure aaron/alexis/jose can provide a short ssh-session for you on their laptop(s)
11:50 jclift ndevos: Yeah, in the Baker St office.  But, there's no spare mouse here and I didn't bring one, so I'm using the touchpad as a mouse today. :)
11:50 ndevos jclift: oh, you really need a mouse?!
11:51 jclift ndevos: Yeah, weird eh? :)
11:51 jclift ndevos: I guess I should just recompile the kernel with psmouse as a module instead, so I can include the ALPS touchpad stuff.
11:52 * jclift grumbles about Fedora not including the stuff.  Works "out of the box" on Ubuntu/Mint/etc.  Just sayin...
11:53 ndevos jclift: if that patch is upstream already, you can send it to kernel@lists.fedoraproject.org and ask if it can be included
11:54 jclift Nah, it's not.  It's still a bit rough, so someone with real kernel coding experience would have to look over it and prob clean it up
11:54 jclift But there's enough users of ALPS touchpads (75%+ of all modern laptops apparently) that Ubuntu/Mint have included it anyway
12:02 edward1 joined #gluster-dev
12:35 kkeithley_ jclift, ndevos: yes, that all sounds good.
12:53 ajha joined #gluster-dev
12:54 kkeithley_ jclift, ndevos: wrt regression test pkg, are you thinking it would be part of the regular glusterfs packaging, or would it be independent of glusterfs and the glusterfs release cycle, sufficiently so, so that it should be in a separate package, as glusterfs-openstack-swift, etc., are?
12:58 ndevos kkeithley_: no, the test-package should be kept with the main glusterfs package - otherwise it'll get messy (tests for features not available etc)
13:01 vshankar joined #gluster-dev
13:05 kkeithley_ that's a good point, I just wanted clarification on that. OTOH I wonder if it wouldn't be just as messy in another way having to maintain multiple sets of tests in different branches for the different releases. If there was a single regression suite and subsets were enabled based on the version of glusterfs being tested, I wonder if that would be easier in some respects. Anyways, it's just a thought. Proceed as you think best.
13:19 ndarshan left #gluster-dev
13:55 awheeler joined #gluster-dev
13:56 awheeler joined #gluster-dev
14:00 hagarth joined #gluster-dev
14:18 wushudoin joined #gluster-dev
14:38 ndevos kkeithley_: a separate regression test suite would be nice, but I think that is not possible yet - maybe when there is a multi-node capable test-suite
14:55 rgustafs joined #gluster-dev
15:05 jclift ndevos kkeithley: Maybe we should adjust the run-tests.sh script so that it sources some test-list.txt file (or similar name), that contains the list of tests to be run for each branch?  eg release-3.4, release-3.3, master
15:06 jclift We'd need some way for run-tests.sh to know which branch it's running in, that I doubt that would be difficult.  Running some git command (eg git branch | cut <something>) would probably extract that reliably if in a git repo.
15:07 jclift s/that I doubt that/but I doubt that/
15:07 ndevos jclift: run-tests.sh will not have a git repository when it is packaged, it would need to execute 'glusterfs --version' or something
15:09 jclift ndevos: Sure.  Same concept applies though.
15:10 jclift a) Figure out what version of gluster we're running.  b) Use that to get the list of tests to run.
15:10 jclift c) Run said tests
15:14 ndevos yes, the concept would work, but I'm not completely clear on the advantage over packaging the versioned tests with the same glusterfs version
15:40 lalatenduM joined #gluster-dev
17:58 hagarth1 joined #gluster-dev
19:00 kkeithley1 joined #gluster-dev
19:09 lpabon joined #gluster-dev
22:27 ndk joined #gluster-dev

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