Perl 6 - the future is here, just unevenly distributed

IRC log for #gluster-dev, 2017-01-31

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

All times shown according to UTC.

Time Nick Message
00:41 nishanth joined #gluster-dev
02:16 shyam joined #gluster-dev
02:51 susant joined #gluster-dev
03:03 susant left #gluster-dev
03:27 rastar joined #gluster-dev
03:33 msvbhat joined #gluster-dev
03:37 magrawal joined #gluster-dev
03:49 itisravi joined #gluster-dev
04:00 susant joined #gluster-dev
04:03 susant left #gluster-dev
04:09 prasanth joined #gluster-dev
04:15 atinm joined #gluster-dev
04:16 skumar joined #gluster-dev
04:26 kdhananjay joined #gluster-dev
04:27 nbalacha joined #gluster-dev
04:31 Saravanakmr joined #gluster-dev
04:41 sanoj joined #gluster-dev
04:42 aravindavk joined #gluster-dev
04:43 mchangir joined #gluster-dev
04:45 rjoseph joined #gluster-dev
04:50 Shu6h3ndu joined #gluster-dev
04:50 Shu6h3ndu joined #gluster-dev
05:05 BlackoutWNCT joined #gluster-dev
05:06 ppai joined #gluster-dev
05:07 ankit_ joined #gluster-dev
05:13 nigelb gasp, no misc.
05:13 susant joined #gluster-dev
05:19 Bluebell_ joined #gluster-dev
05:21 ndarshan joined #gluster-dev
05:30 BlackoutWNCT joined #gluster-dev
05:38 apandey joined #gluster-dev
05:45 gyadav joined #gluster-dev
05:48 BlackoutWNCT joined #gluster-dev
05:49 BlackoutWNCT joined #gluster-dev
05:49 BlackoutWNCT joined #gluster-dev
05:50 vimal joined #gluster-dev
05:53 riyas joined #gluster-dev
06:02 msvbhat joined #gluster-dev
06:03 BlackoutWNCT joined #gluster-dev
06:03 asengupt joined #gluster-dev
06:04 BlackoutWNCT joined #gluster-dev
06:04 BlackoutWNCT joined #gluster-dev
06:10 rafi joined #gluster-dev
06:14 skoduri joined #gluster-dev
06:15 poornima joined #gluster-dev
06:33 msvbhat joined #gluster-dev
06:43 sanoj joined #gluster-dev
06:46 apandey joined #gluster-dev
06:46 nishanth joined #gluster-dev
06:53 jiffin joined #gluster-dev
06:59 BlackoutWNCT joined #gluster-dev
07:00 BlackoutWNCT joined #gluster-dev
07:00 msvbhat joined #gluster-dev
07:12 Humble joined #gluster-dev
07:13 rraja joined #gluster-dev
07:21 misc joined #gluster-dev
08:00 nbalacha joined #gluster-dev
08:40 nbalacha joined #gluster-dev
08:40 Bluebell_ left #gluster-dev
08:48 gem joined #gluster-dev
08:58 atinm xavih, are you there?
08:59 atinm xavih, this is regarding the op-version issue you highlighted
09:07 xavih atinm: hi
09:16 atinm xavih, forward porting is not a problem for op-version management, its the back porting which will cause issues
09:17 xavih atinm: I know, but if I create a new option for 3.7.21, it will be considered valid for 3.8.0, right ?
09:17 atinm xavih, w.r.t LTS, your concern is valid, if users do not want to upgrade to a short term release then we do have such problems and they have to wait till the next LTS comes in
09:17 itisravi joined #gluster-dev
09:18 atinm xavih, the compatibility check will go through but the option will not exist in 3.8.0 is what my biggest concern is
09:18 xavih atinm: yes, 3.8.0 will fail because of an unknown option
09:19 xavih atinm: some options are not critical if unknown, so they could be simply ignored
09:19 atinm kshlm, any thoughts on this?
09:20 kshlm IIRC we've had to do this before.
09:20 xavih atinm: that would solve the problem of adding new options that need to be backported
09:20 kshlm The options was given the op-version of the oldest version it would be backported to.
09:20 atinm kshlm, how did we do it?
09:20 kshlm And we would ask the users to only upgrade to the latest new version.
09:21 kshlm For eg. if the options went into (say) 3.8.50 and 3.9.20
09:21 kshlm we'd only support upgrades from 3.8.50 to 3.9.20 and newer versions.
09:22 kshlm There is no way for us to enforce that updates happen this way.
09:22 kshlm But we should document this, and make it well known.
09:25 xavih kshlm: I don't understand... so the option will be defined with GD_OP_VERSION_3_9_20 ?
09:25 xavih kshlm: how will this option be allowed by a 3.8.50 version ?
09:25 kshlm xavih, The op-version will be 3.8.50
09:25 kshlm That is the oldest version you are backporting to.
09:25 xavih kshlm: ah, ok
09:27 xavih kshlm: then the only control would be by the user itself. The user must make sure he upgrades to the correct versions
09:28 kshlm Yup.
09:28 kshlm That's how I believe other applications work as well.
09:28 kotreshhr joined #gluster-dev
09:29 xavih kshlm: does this work well with the brick multiplexing ? this feature will force a massive update of all volumes and clients each time a new option is defined...
09:30 kshlm Not unless you would need to enable the feature for all volumes.
09:31 kshlm Even with brick multiplexing, we have individual volfiles for each brick.
09:31 pkalever joined #gluster-dev
09:31 kshlm So changing an option on a volume will only affect that volume.
09:32 xavih kshlm: I see
09:33 * kshlm is having doubts and will need to check this part of multiplexing again.
09:35 xavih kshlm: ok, so if I want to create a new option that needs to be backported, I must create it for the older next version and document it so the users be aware of the upgrade restrictions, right ?
09:35 kshlm That is correct.
09:35 xavih kshlm: where should this be documented ? in the release notes ?
09:35 kshlm Yup.
09:35 kshlm That would be in the right place.
09:36 xavih kshlm: ok, thanks for the information :)
09:36 kshlm Oh wow. The multiplexing review got merged! Awesome.
09:36 kshlm xavih, You're welcome.
09:36 xavih kshlm++
09:36 glusterbot xavih: kshlm's karma is now 136
09:36 xavih atinm++
09:36 glusterbot xavih: atinm's karma is now 75
09:37 gyadav joined #gluster-dev
09:38 nbalacha joined #gluster-dev
10:22 nbalacha joined #gluster-dev
10:22 ShwethaHP joined #gluster-dev
10:52 pkalever joined #gluster-dev
10:59 pkalever left #gluster-dev
11:03 nbalacha joined #gluster-dev
11:11 poornima skoduri++
11:11 glusterbot poornima: skoduri's karma is now 52
11:22 nbalacha joined #gluster-dev
11:24 ankit__ joined #gluster-dev
11:26 ankit_ joined #gluster-dev
11:26 gyadav_ joined #gluster-dev
11:28 shyam joined #gluster-dev
11:33 kkeithley bug triage meeting in approx 30 minutes in #gluster-meeting
11:34 gem joined #gluster-dev
11:59 kkeithley bug triage meeting now in #gluster-meeting
12:13 kotreshhr left #gluster-dev
12:33 ashiq joined #gluster-dev
12:35 jiffin joined #gluster-dev
12:47 ndarshan joined #gluster-dev
13:00 rraja joined #gluster-dev
13:06 ira joined #gluster-dev
13:19 kotreshhr joined #gluster-dev
13:28 ppai kshlm, #230 is easy review. You can rebase #225 against it in your next PR update. After this, I hope to not send PRs that conflict with #225 until it gets merged.
13:35 shyam joined #gluster-dev
13:38 Saravanakmr joined #gluster-dev
13:49 atinm joined #gluster-dev
14:07 pkalever joined #gluster-dev
14:12 kotreshhr left #gluster-dev
14:26 msvbhat joined #gluster-dev
14:34 msvbhat joined #gluster-dev
14:35 nbalacha joined #gluster-dev
14:53 ankit_ joined #gluster-dev
15:18 atinm joined #gluster-dev
15:21 susant joined #gluster-dev
15:22 pkalever joined #gluster-dev
15:29 ndevos shyam: have you heard anything back from poornima on https://review.gluster.org/16357 ?
15:30 shyam ndevos: nope!
15:30 ndevos :-(
15:30 shyam I have some questions for you Niels :) on the bug list in the release notes, if you have a few minutes...
15:30 ndevos shyam: I'm not sure if she expects me to update the spec too...
15:30 ndevos shyam: sure
15:31 gyadav_ joined #gluster-dev
15:31 shyam ok on the bug list, when using the release_notes.sh to generate the list, there are a bunch of false positives, i.e things fixed in 3.9 but now appearing against the 3.10 list
15:31 ndevos rjoseph: do you know if poornima was planning to update the design/spec for the gfapi statedump feature?
15:31 shyam ndevos: Did you cull the list post the output. when doing 3.8 stuff?
15:32 shyam Also, what if a bug is fixed in 3.9.1 should it still appear in the 3.10 release notes?
15:32 ndevos shyam: no, I kept it as is, those are still bugs that are fixed in the release
15:32 shyam yes, the list is inclusive, so that way it is not a problem
15:32 ndevos they are different bugs right, each bug has a different version
15:32 shyam But it is larger than what it should be, so wondering if we should reduce it
15:33 shyam yes
15:33 ppai kshlm++
15:33 glusterbot ppai: kshlm's karma is now 137
15:33 ndevos you could reduce it, check for each bug if it was cloned for/from a 3.9 one and if that has been closed... but that may be a lot of work
15:33 shyam yup it is :)
15:34 ndevos I try to have the correct dependencies on bugs, but there is no automation and many people that clone bugs have the dependency backwards
15:34 shyam ok, so this is not new, let me think about it. I was wondering if I was doing something incorrect, appears not
15:35 ndevos I always though it would not hurt to mention all the bugs that were fixed in the release
15:35 ndevos *thought
15:36 ndevos we also do not expect users to move only from 3.9 to 3.10, also 3.7 users will be interested and they still want to know all bugs that were fixed
15:36 shyam yeah that is what I was thinking as well... but it would be nice to have an actual set and possibly rules that state, this is a set of bugs post 3.9.0 release that are fixed in 3.10.0 and possibly are part of further 3.9 minor releases (change 9/10 to anything)
15:36 ndevos although the 3.7 users will have a gap in the bug listing...
15:36 shyam yes, the current script is last -> now
15:36 shyam Which brings another interesting question should the list for 3.10 be from 3.8 or form 3.9 (STM/LTM)
15:37 ndevos yeah, I was just thinking about that too
15:37 shyam (well current script is not last->now, it's just how I used it, for the record)
15:37 ndevos probably listing all the bugs from 3.8->3.10 would be good
15:37 ndevos or point to the 3.9 release notes
15:38 shyam Because, using the script we should be able to give it a From TAG till a to TAG, so we can do things like, post last 3.8.x release till 3.10.0
15:38 shyam ok something to chew on, let me generate the lists and see what makes sense
15:38 ndevos yes, just list it from the 1st commit in master after the branching of 3.8 (or 3.9)
15:40 shyam above + post branching 3.10 till day before release, should cover additional backports to 3.10
15:40 ndevos https://github.com/gluster/glusterfs/blo​b/release-3.9/doc/release-notes/3.9.0.md lists 422 bugs, it may be a little much to include all of those too
15:42 ndevos the history from the 1st commit in master after 3.9 branching until the HEAD of release-3.8 is linear, the script should be able to handle that just fine
15:42 shyam the current count when using the same script as `release_notes.sh release-3.9 release-3.10 <path to repo>` is 274
15:45 msvbhat joined #gluster-dev
15:48 kkeithley did something just change on download.gluster.org?  wget from debian wheezy box (i.e. old) is suddenly complaining that it doesn't trust the d.g.o certificate
15:48 kkeithley suddenly as in since the last time I did a wget on this box, around Jan 1 or so
15:48 kkeithley nigelb, misc: ^^^
15:51 kkeithley apt says i have the latest ca-certificates pkg installed
15:52 lpabon joined #gluster-dev
15:54 timotheus1 joined #gluster-dev
15:58 misc mhh ?
16:00 misc debian wheezy is old stable ?
16:00 misc (cause it work on debian 8.6)
16:04 wushudoin joined #gluster-dev
16:07 wushudoin joined #gluster-dev
16:10 gyadav joined #gluster-dev
16:19 msvbhat joined #gluster-dev
16:26 pkalever left #gluster-dev
16:28 ndevos nigelb: this got a weird error at the end of a smoke test, I've never seen that before, you? https://build.gluster.org/job/smoke/33115/console
16:28 susant left #gluster-dev
16:30 ndevos actually, that might be a dbench failure again, I thought that problem got resolved?
16:55 gyadav joined #gluster-dev
17:07 ankit_ joined #gluster-dev
17:31 riyas joined #gluster-dev
17:47 zeio joined #gluster-dev
17:52 ppai joined #gluster-dev
17:57 msvbhat joined #gluster-dev
18:13 ShwethaHP joined #gluster-dev
18:20 vbellur joined #gluster-dev
18:22 msvbhat joined #gluster-dev
18:53 msvbhat joined #gluster-dev
19:04 msvbhat joined #gluster-dev
19:27 msvbhat joined #gluster-dev
19:40 msvbhat joined #gluster-dev
19:46 vbellur joined #gluster-dev
20:06 ankit_ joined #gluster-dev
20:09 vbellur joined #gluster-dev
20:45 vbellur joined #gluster-dev
20:52 jiffin joined #gluster-dev
21:02 ashiq joined #gluster-dev
22:09 vbellur joined #gluster-dev
22:40 kkeithley joined #gluster-dev

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