--- Log opened di okt 07 00:00:52 2014 00:01 -!- adahms (Andrew Dahms): has joined #vdsm 00:52 -!- fsimonce: has quit [Quit: Coyote finally caught me] 00:55 -!- #vdsm jbrooks: has quit [Ping timeout: 240 seconds] 01:14 -!- jbrooks (Jason Brooks): has joined #vdsm 01:18 -!- #vdsm jbrooks: has quit [Ping timeout: 245 seconds] 01:21 -!- jbrooks (Jason Brooks): has joined #vdsm 01:55 -!- adahms: has quit [Quit: Leaving] 01:59 -!- #vdsm jbrooks: has quit [Ping timeout: 272 seconds] 02:02 -!- adahms (Andrew Dahms): has joined #vdsm 02:03 -!- jbrooks (Jason Brooks): has joined #vdsm 02:47 -!- #vdsm jbrooks: has quit [Ping timeout: 250 seconds] 02:49 -!- jbrooks (Jason Brooks): has joined #vdsm 03:09 -!- adahms: has quit [Ping timeout: 260 seconds] 03:09 -!- moolit: has quit [Quit: Leaving.] 03:16 -!- #vdsm jbrooks: has quit [Ping timeout: 246 seconds] 03:21 -!- adahms (Andrew Dahms): has joined #vdsm 03:22 -!- nsoffer: has quit [Ping timeout: 260 seconds] 03:31 -!- jbrooks (Jason Brooks): has joined #vdsm 06:04 -!- fabiand (Fabian Deutsch): has joined #vdsm 06:06 -!- vered (Vered Volansky): has joined #vdsm 06:16 -!- vered: has quit [Quit: Leaving] 06:19 -!- vered (Vered Volansky): has joined #vdsm 06:58 -!- bala (purple): has joined #vdsm 06:59 -!- oved (ovedo): has joined #vdsm 07:35 -!- gpadgett: has quit [Ping timeout: 260 seconds] 07:39 -!- ishaby (Idan Shaby): has joined #vdsm 07:55 -!- sbonazzo (purple): has joined #vdsm 08:00 -!- fromani (Francesco Romani): has joined #vdsm 08:05 -!- mskrivanek_away is now known as mskrivanek 08:09 -!- #vdsm ybronhei: has quit [Quit: Leaving.] 08:12 -!- dkuznets (Dima Kuznetsov): has joined #vdsm 08:37 -!- pkliczew (Piotr Kliczewski): has joined #vdsm 08:45 -!- ishaby: has quit [Quit: Leaving] 08:45 -!- ishaby (Idan Shaby): has joined #vdsm 08:46 -!- xaviern (Xavier): has joined #vdsm 08:53 -!- ybronhei (purple): has joined #vdsm 08:54 -!- #vdsm xaviern: has quit [Ping timeout: 245 seconds] 08:57 -!- adahms: has quit [Quit: Leaving] 08:58 -!- #vdsm ybronhei: has quit [Ping timeout: 240 seconds] 09:02 -!- adahms (Andrew Dahms): has joined #vdsm 09:03 -!- adahms: has quit [Client Quit] 09:07 -!- mpolednik (Martin Polednik): has joined #vdsm 09:26 -!- xaviern (Xavier): has joined #vdsm 09:32 -!- fsimonce (Federico): has joined #vdsm 09:34 -!- acanan (Aharon Canan): has joined #vdsm 09:46 -!- ybronhei (purple): has joined #vdsm 09:57 -!- msivak (Martin Sivák): has joined #vdsm 10:00 -!- phoracek (phoracek): has joined #vdsm 10:12 -!- adahms (Andrew Dahms): has joined #vdsm 10:20 -!- mode/#vdsm: by ChanServ 10:20 -!- danken (purple): has joined #vdsm 10:20 <@danken> pkliczew: http://gerrit.ovirt.org/#/c/33861/1/vdsm/rpc/Bridge.py seems urgent 10:21 < pkliczew> danken, yes 10:21 <@danken> pkliczew: what's the functional effect? 10:21 <@danken> oh 10:21 <@danken> Bug 1148499 - : detachStorageDomain fails with StoragePoolUnknown for destroying a data domain 10:21 <@danken> ? 10:21 <@danken> only that? 10:21 < pkliczew> danken, force detach fails 10:21 < pkliczew> danken, on any domain 10:22 < pkliczew> danken, we haven't seen this issue occurring for other flows so far 10:22 <@danken> is that because of unicode, by any chance? 10:22 < pkliczew> not sure 10:22 <@danken> (class_name == className) is just like (class_name is className) 10:22 <@danken> for strings 10:23 < pkliczew> it works for Host but return false for StorageDomain 10:24 < phoracek> danken: hi, is there a way to destroy persisted networks in vdsm? when i run test, it returns EnvironmentError: Failed to restore the persisted networks 10:25 <@danken> phoracek: you can kill the persistent data by removing the dir from /var/run/vdsm/,,, 10:25 < phoracek> danken: thanks! :) 10:25 <@danken> phoracek: but it should be interesting to debug that error, too 10:25 <@danken> because it should not happen 10:26 < phoracek> ok, i'm gonna do little investigation 10:26 <@danken> phoracek: clearly, we had a bug persisting something that we should not (if it's an empty bond device, Toni has already solved that) 10:27 < evilissimo1> fromani: http://gerrit.ovirt.org/#/c/33881/ 10:27 < fromani> evilissimo1: just ACKed 10:27 < evilissimo1> fromani: that sucks, right? 10:27 < fromani> evilissimo1: too bad I haven't noticed in the original review 10:27 -!- evilissimo1 is now known as evilissimo 10:28 < fromani> evilissimo: the issue yes, the fix looks nice :) 10:28 < evilissimo> lol nice 10:28 < fromani> we need more comprehensive recovery tests :\ 10:28 <@danken> pkliczew: I still like to understand why you think "is" is wrong. 10:29 <@danken> (I had my own guess, but I wonder what made you do this change) 10:29 < pkliczew> danken, I tested it and saw it failing, give me some time and I will check exactly by reproducing it again 10:30 < evilissimo> danken: what is, if unicodestring vs string? 10:30 <@danken> evilissimo: yeah, that's my guess. 10:30 < evilissimo> that gives you false 10:31 < evilissimo> danken: but "foo" == u"foo" is fine 10:32 < evilissimo> To be honest when dealing with strings I would not go for 'is' 10:32 < evilissimo> at least not in python2 10:33 <@danken> evilissimo: yet 10:33 <@danken> "foo" is u"foo" == False 10:34 < evilissimo> danken: of course, because they are not of the same type 10:34 < evilissimo> 1 is 1L == False 10:34 < evilissimo> same thing 10:35 < evilissimo> however 1 == 1L yields True 10:36 < sbonazzo> danken: good morning, do you have ETA on 3.5.0 blocker Bug 1149705 - VM abnormal stop after LV refreshing when using thin provisioning on block storage ? 10:36 -!- apahim: has quit [Ping timeout: 246 seconds] 10:39 < evilissimo> danken: I was just talking with msivak about it and he made a pretty good point I did not consider as argument: "is" only check for identity 10:39 -!- moolit (purple): has joined #vdsm 10:40 < evilissimo> >>> a = 1022 10:40 < evilissimo> >>> b = 1012 + 10 10:40 < evilissimo> >>> a is b 10:40 < evilissimo> False 10:40 < evilissimo> danken 10:41 -!- apuimedo (antoni): has joined #vdsm 10:41 <@danken> sbonazzo: huh?! I did not know it was reopened. fsimonce? 10:41 < fromani> wasn't "is" definied as "a is b if and only if id(a) == id(b)"? 10:42 <@danken> sbonazzo: Bug 1148499 - : detachStorageDomain fails with StoragePoolUnknown for destroying a data domain 10:42 <@danken> should be a blocker, too. IMO 10:42 < fsimonce> danken, the patch was reverted... nsoffer is working on it 10:42 < sbonazzo> danken: feel free to add it to the tracker 10:42 < fsimonce> danken, we also have an email thread 10:42 <@danken> fsimonce: oh, Ihaven't read my email yet 10:43 < fsimonce> danken, we'll probably be abandoning the udev permissions 10:43 < evilissimo> danken: and about strings: >>> s = raw_input() 10:43 < evilissimo> foobar 10:43 < evilissimo> >>> s2 = raw_input() 10:43 < evilissimo> foobar 10:43 < evilissimo> >>> s is s2 10:43 < evilissimo> False 10:44 < evilissimo> >>> s == s2 10:44 < evilissimo> True 10:44 < evilissimo> sorry for spamming too much :S 10:44 < evilissimo> danken: both s and s2 are of type str 10:45 < evilissimo> pkliczew: ^ 10:45 < pkliczew> evilissimo, ok 10:45 < evilissimo> pkliczew: so always use == for string comparison 10:46 < evilissimo> or value comparison 10:46 < evilissimo> to be more precise 10:46 < pkliczew> evilissimo, yes == compares values and is compare identity (object) 10:46 <@danken> fsimonce: on systemd side? I understood that from nsoffer email from sunday 10:46 < evilissimo> pkliczew: yes 10:48 <@danken> evilissimo: this surprises me a lot 10:48 <@danken> live and learn 10:48 < evilissimo> danken: well good to have a python guru around here :-) 10:48 < pkliczew> evilissimo, danken in this case 'Host' string literal is defined in _convertClassName so when compared is the same whereas it is the same object whereas any other className is concatenated from a message so are different objects 10:48 < pkliczew> evilissimo, :-) 10:48 < evilissimo> danken: With that I am refering to msivak ;) 10:48 -!- ykaplan (Yeela Kaplan): has joined #vdsm 10:49 < fsimonce> danken, the email I sent yesterday (it works on all platforms, el6, el7, fedora*) 10:54 -!- mskrivanek is now known as mskrivanek_away 10:55 -!- mskrivanek_away is now known as mskrivanek 11:01 -!- moolit: has quit [Ping timeout: 272 seconds] 11:03 -!- phoracek: has quit [Quit: WeeChat 0.4.3] 11:04 -!- phoracek (phoracek): has joined #vdsm 11:08 <@danken> pkliczew: I took your master patch; 3.5 is next 11:08 < pkliczew> danken, already pushed 11:08 < pkliczew> danken, http://gerrit.ovirt.org/#/c/33882/ 11:11 <@danken> ybronhei: it's all yours ^^^ 11:13 -!- wgao: has quit [Quit: Leaving] 11:15 -!- moolit (purple): has joined #vdsm 11:18 -!- osvoboda (osvoboda): has joined #vdsm 11:23 -!- fromani_ (Francesco Romani): has joined #vdsm 11:23 -!- fromani_: has quit [Client Quit] 11:24 -!- fromani: has quit [Ping timeout: 245 seconds] 11:24 -!- fromani: has quit [Client Quit] 11:35 -!- osvoboda: has quit [Quit: ChatZilla 0.9.91 [Firefox 32.0.3/20140925003805]] 11:38 < apuimedo> osvoboda: meh, the patch that adds the new bonding configuration file makes me a bit sad 11:38 <@danken> sbonazzo: fsimonce: I see that Bug 1149705 - VM abnormal stop after LV refreshing when using thin provisioning on block storage 11:38 <@danken> is MODIFIED again 11:38 < apuimedo> I used to be able to run PYTOHNPATH="lib:$PYTHONPATH" ipython 11:38 < apuimedo> and then 11:38 < apuimedo> from vdsm import vdscli 11:38 < apuimedo> sorry 11:39 < apuimedo> from vdsm import netinfo 11:39 < apuimedo> and be able to get netinfo.get() 11:39 < sbonazzo> danken: so, we need a rebuild of vdsm and engine 11:39 <@danken> sbonazzo: there was not vdsm-side change. I am not sure that I follow what's going on there. 11:41 <@danken> sbonazzo: but we need another build, anyway for pkliczew's http://gerrit.ovirt.org/#/c/33882/ 11:42 < sbonazzo> danken: ok, we've 4 patches on ovirt 3.5.0 branch, trying to understand right now if we need all of them 11:45 -!- ykaplan: has quit [Ping timeout: 246 seconds] 11:47 <@danken> 4 patches? which? 11:47 <@danken> I know only of pkliczew's. 11:50 <@danken> apuimedo: we can move the in-git copy into the lib/vdsm package 11:50 <@danken> and convert it from json to py 11:51 < apuimedo> danken: are you advocating for code generation? 11:52 < apuimedo> I find it agreeable 11:52 < apuimedo> we'll have to test that selinux doesn't play nasty games with the imports if the user re-generates it with vdsm-tool 11:55 < sbonazzo> danken: on engine: http://gerrit.ovirt.org/#/q/status:open+project:ovirt-engine+branch:ovirt-engine-3.5.0,n,z 11:59 < ybronhei> danken: pkliczew: why == better there than 'is' ? 12:00 < ybronhei> danken: pkliczew http://gerrit.ovirt.org/#/c/33882/2/vdsm/rpc/Bridge.py,cm ^ 12:00 < pkliczew> ybronhei, above there is discussion about it 12:01 < ybronhei> i look for it .. 12:01 -!- sbonazzo is now known as sbonazzo|afk 12:01 < ybronhei> pkliczew: where above? 12:01 < pkliczew> ybronhei, on this channel -> evilissimo, danken in this case 'Host' string literal is defined in _convertClassName so when compared is the same whereas it is the same object whereas any other className is concatenated from a message so are different objects 12:02 <@danken> ybronhei: if one of the sides is unicode u'bla' 12:02 <@danken> then the "is" fails 12:02 <@danken> equality succeeds 12:03 <@danken> also , vinzenz reported of cases were str-str fails with "is" too. 12:03 <@danken> it surprised me 12:04 <@danken> fromani: mskrivanek: ? 12:04 < ybronhei> so the conclusion is that better to use == then is for strings 12:05 < fromani> danken: i'm here 12:05 < ybronhei> anyhow, regarding to rhel7 issue - i see in the spec that we don't require libvirt in rhel 7 and fedora > 18 12:06 < ybronhei> only libvirt-daemon 12:06 < ybronhei> should it be that way danken? 12:07 < ybronhei> danken: other than that, i saw that there are some issues without the z-stream repo for rhel7 . should we try to reproduce it to fine the specific broken package? after yum upgrade it works .. 12:14 -!- osvoboda (osvoboda): has joined #vdsm 12:16 < ybronhei> apuimedo: i have a question , do you have a minute ? 12:16 < ybronhei> apuimedo: even 10 12:22 < fromani> another RHEL 6.6 libvirt blocker for us https://bugzilla.redhat.com/show_bug.cgi?id=1149667 12:22 -!- ykaplan (Yeela Kaplan): has joined #vdsm 12:26 < apuimedo> ybronhei: i do 12:28 < ybronhei> apuimedo: i think i figured that out .. but not sure 12:28 < apuimedo> ok 12:28 < ybronhei> apuimedo: network issue with named 12:32 <@danken> mpolednik: can you confirm that http://gerrit.ovirt.org/#/c/32313/ flies with jsonrpc? 12:33 <@danken> I'm adding pkliczew anyway, he's got a good eyes for things that breaks API. 12:33 -!- osvoboda-mobile (Ondřej Svoboda): has joined #vdsm 12:34 -!- osvoboda-mobile2 (Ondřej Svoboda): has joined #vdsm 12:35 -!- osvoboda-mobile2: has quit [Client Quit] 12:38 -!- ishaby: has quit [Remote host closed the connection] 12:39 -!- mskrivanek is now known as mskrivanek_away 12:40 < mpolednik> danken: what would be the routine to actually test json via available tools? 12:45 < evilissimo> ybronhei: the conclusion is that you use == when you want to compare content 12:46 < evilissimo> ybronhei: 'is' compares identity, the point why 'str' == 'str' in the interpreter results in True is, that the code gets 'compiled' to byte code and the interpreter sees two identical strings and creates only one real object for it 12:46 -!- ishaby (Idan Shaby): has joined #vdsm 12:46 < evilissimo> ybronhei: however if you read something on runtime it's creating a new string object every time where the identity is not the same 12:48 < evilissimo> imagine this in C: char const foobar[] = "foobar"; char const * foobar2 = malloc(sizeof(foobar)); strcpy(foobar2, foobar); strncmp(foobar, foobar2, sizeof(foobar) -1) == 0 => True, foobar == foobar2 => False 12:48 < evilissimo> that's comparable 12:48 < ybronhei> right 12:51 < pkliczew> saggi, ping 12:52 < evilissimo> ybronhei: c++ terms would have been easier to show off :D, std::string s = "foobar", s2 = "foobar"; s == s2 => True; &s == &s2 => False xD 12:52 -!- apahim (Amador Pahim): has joined #vdsm 12:53 < ybronhei> so 'is' checks the pointer, that's what you say 12:53 < evilissimo> that's a naive way of looking at it 12:53 < evilissimo> but you could say so 12:54 -!- Netsplit *.net <-> *.split quits: bala, vered, Humble, rmatinata 12:54 -!- Humble (Humble Chirammal): has joined #vdsm 12:55 < evilissimo> basically the pointer can refer to its identity since no two objects can have theoretically the same address. In practice it's a bit more difficult, at least when you again look at C, in C it's quite possible that two different objects have the same address 12:55 < evilissimo> one is the address to a structure the other one to the first member in the structure 12:55 < evilissimo> and still they would have the same address 12:56 < evilissimo> s/would/could/ 12:56 < evilissimo> that's then a question of memory layout etc 12:59 <@danken> mpolednik: oh, we do not have a command line to test that 12:59 <@danken> and you don't have Engine-side support for your new verb either 13:00 <@danken> I think you are left with nagging to pkliczew for his review 13:00 <@danken> mpolednik: also, you should say HOW you verified your patch when you set V+1. 13:01 < mpolednik> danken: same way as in the previous v+1, will be more verbose later 13:06 -!- saggi: has quit [Ping timeout: 244 seconds] 13:07 -!- vered (Vered Volansky): has joined #vdsm 13:07 -!- rmatinata (Ricardo Marin Matinata): has joined #vdsm 13:09 -!- fromani_ (Francesco Romani): has joined #vdsm 13:10 -!- fromani: has quit [Ping timeout: 245 seconds] 13:17 -!- pkliczew: has quit [Ping timeout: 245 seconds] 13:20 -!- pkliczew (Piotr Kliczewski): has joined #vdsm 13:21 -!- fromani_ is now known as fromani 13:21 -!- fromani: has quit [Quit: Leaving] 13:34 -!- mpolednik: has quit [Ping timeout: 260 seconds] 13:35 -!- mpolednik (Martin Polednik): has joined #vdsm 13:36 -!- saggi (purple): has joined #vdsm 13:43 -!- apuimedo: has quit [Ping timeout: 245 seconds] 13:46 -!- sbonazzo|afk is now known as sbonazzo 13:54 -!- osvoboda-mobile2 (Ondřej Svoboda): has joined #vdsm 13:54 -!- osvoboda-mobile: has quit [Ping timeout: 272 seconds] 13:54 < osvoboda-mobile2> danken: I will check my new patch after a lunch. I hope it conforms to your idea :-) 14:01 -!- osvoboda: has quit [Ping timeout: 245 seconds] 14:05 -!- vered: has quit [Ping timeout: 260 seconds] 14:05 -!- osvoboda-mobile2: has quit [Ping timeout: 244 seconds] 14:11 -!- osvoboda-mobile2 (Ondřej Svoboda): has joined #vdsm 14:12 -!- mpolednik: has quit [Ping timeout: 244 seconds] 14:17 <@danken> osvoboda-mobile2: let toni review it - it is his own peeve 14:27 -!- #vdsm ybronhei: has quit [Ping timeout: 250 seconds] 14:28 -!- moolit: has quit [Ping timeout: 250 seconds] 14:32 -!- oved: has quit [Ping timeout: 272 seconds] 14:33 -!- acanan_ (Aharon Canan): has joined #vdsm 14:34 -!- acanan: has quit [Ping timeout: 245 seconds] 14:41 -!- ybronhei (purple): has joined #vdsm 14:44 -!- moolit (purple): has joined #vdsm 14:47 -!- oved (ovedo): has joined #vdsm 14:55 -!- nsoffer (Nir Soffer): has joined #vdsm 14:58 -!- ykaplan: has quit [Ping timeout: 250 seconds] 14:59 -!- mpolednik (Martin Polednik): has joined #vdsm 15:02 -!- #vdsm ybronhei: has quit [Ping timeout: 258 seconds] 15:07 -!- mskrivanek_away is now known as mskrivanek 15:08 -!- ykaplan (Yeela Kaplan): has joined #vdsm 15:11 -!- moolit: has quit [Ping timeout: 246 seconds] 15:11 -!- rmatinata: has quit [Ping timeout: 249 seconds] 15:15 -!- gpadgett (Greg Padgett): has joined #vdsm 15:17 -!- ybronhei (purple): has joined #vdsm 15:25 -!- moolit (purple): has joined #vdsm 15:47 -!- apuimedo (antoni): has joined #vdsm 15:48 -!- dyasny (Dan Yasny): has joined #vdsm 15:48 < apuimedo> danken: did you see the bug/flame in https://bugzilla.gnome.org/725647 15:48 < apuimedo> ? 15:52 -!- Thread-65536 (Android SimpleIrc): has joined #vdsm 15:56 -!- #vdsm xaviern: has quit [Quit: WeeChat 0.4.3] 15:56 -!- Thread-65536: has quit [Remote host closed the connection] 15:57 -!- xaviern (Xavier): has joined #vdsm 15:58 -!- mskrivanek is now known as mskrivanek_away 16:03 -!- mpolednik: has quit [Quit: leaving] 16:03 -!- osvoboda (osvoboda): has joined #vdsm 16:03 -!- rmatinata (Ricardo Marin Matinata): has joined #vdsm 16:05 -!- mskrivanek_away is now known as mskrivanek 16:06 -!- osvoboda-mobile3 (Ondřej Svoboda): has joined #vdsm 16:08 -!- osvoboda-mobile2: has quit [Read error: Connection reset by peer] 16:11 -!- ishaby: has quit [Ping timeout: 250 seconds] 16:19 -!- adahms: has quit [Quit: Leaving] 16:29 -!- ykaplan: has quit [Ping timeout: 245 seconds] 16:47 -!- oved: has quit [Ping timeout: 272 seconds] 16:48 -!- ykaplan (Yeela Kaplan): has joined #vdsm 16:57 -!- #vdsm xaviern: has quit [Ping timeout: 258 seconds] 17:05 -!- pkliczew: has quit [Remote host closed the connection] 17:11 -!- sbonazzo: has quit [Quit: Leaving.] 17:12 -!- acanan_: has quit [Remote host closed the connection] 17:15 -!- moolit: has quit [Ping timeout: 244 seconds] 17:19 -!- xaviern (Xavier): has joined #vdsm 17:27 -!- acanan (Aharon Canan): has joined #vdsm 17:39 -!- #vdsm xaviern: has quit [Ping timeout: 245 seconds] 17:47 -!- fabiand: has quit [Quit: Verlassend] 17:48 -!- ykaplan: has quit [Remote host closed the connection] 17:52 -!- #vdsm ybronhei: has quit [Ping timeout: 240 seconds] 17:52 -!- dkuznets: has quit [Ping timeout: 244 seconds] 17:56 < nsoffer> apuimedo, ping 17:56 < apuimedo> nsoffer: pong 17:56 < nsoffer> apuimedo, I have networking issue with nested kvm vms running vdsm 17:57 < nsoffer> apuimedo, from time to time, my vms loose the ip address 17:57 < nsoffer> all machines are running master 17:57 < apuimedo> nsoffer: dhcp? 17:57 < apuimedo> rhel7? 17:57 < nsoffer> all using dhcp in tlv 17:57 < nsoffer> static dhcp 17:57 < nsoffer> e.g. voodoo6.tlv.redhat.com 17:58 < nsoffer> I have seen this on rhel7 and fedora 20 17:58 < apuimedo> nsoffer: it's a bug in dhclient-script 17:58 < nsoffer> when this happens, running systemctl restart network fixes it 17:58 < apuimedo> nsoffer: we asked for a backport 17:58 < apuimedo> f21 and el7.1 are scheduled to get it 17:59 < apuimedo> but we requested it for el7.0 too 17:59 < apuimedo> nsoffer: I can point you to the necessary change in /usr/sbin/dhclient-script if you want 17:59 < apuimedo> in the meantime 17:59 < nsoffer> apuimedo, that would be great 18:01 < apuimedo> nsoffer: trying to find it after a few days 18:08 -!- moolit (purple): has joined #vdsm 18:24 -!- moolit: has quit [Quit: Leaving.] 18:26 -!- fromani: has quit [Quit: Leaving] 18:28 -!- acanan: has quit [Ping timeout: 272 seconds] 18:42 -!- phoracek: has quit [Ping timeout: 260 seconds] 18:51 -!- osvoboda: has quit [Ping timeout: 246 seconds] 18:52 -!- osvoboda-mobile3: has quit [Ping timeout: 246 seconds] 18:54 -!- mskrivanek is now known as mskrivanek_away 18:56 -!- moolit (purple): has joined #vdsm 19:25 < apuimedo> danken: do you remember where do we define the iptables rules for vdsm? 19:26 < apuimedo> oh, it's the vdsm bootstrap still 19:29 -!- phoracek (phoracek): has joined #vdsm 19:46 -!- phoracek: has quit [Quit: WeeChat 0.4.3] 19:50 -!- osvoboda (osvoboda): has joined #vdsm 20:04 <@danken> apuimedo: nowadays it is passed from Engine via ovirt-host-deploy 20:05 < apuimedo> danken: I saw 20:05 < apuimedo> I'm discussing it very intensely with alonbl 20:06 < apuimedo> I just submitted patches to firewalld 20:06 < apuimedo> to have a default service upstream 20:06 < apuimedo> called vdsm.xml 20:06 < apuimedo> that has what we need 20:06 < apuimedo> alonbl prefers to generate them 20:07 <@danken> genereate from what? 20:08 < apuimedo> from some templates he made 20:08 < apuimedo> in otopi 20:08 <@danken> it sounds like something that should be shipped by vdsm.rpm 20:08 < apuimedo> danken: nah 20:08 < apuimedo> in firewalld rpm 20:08 < apuimedo> only in the vdsm rpm if we wanted to reduce the risk of depending to specific firewalld versions 20:09 < apuimedo> if we think that we are going to be changing ports often 20:11 <@danken> it reminds me of the great debate about whether each app should ship with its own selinux context 20:12 <@danken> so I see the security point in deploying the rules in firewalld 20:12 < apuimedo> danken: first I checked how many services they have 20:12 < apuimedo> saw that they have ever tor 20:13 < apuimedo> and said 20:13 < apuimedo> ok, we can be there too 20:14 < apuimedo> I'm arguing for just writing to /etc those that are not default config 20:14 < apuimedo> but alonbl prefers to just generate from templates and write them in /etc as ovirt- 20:15 < apuimedo> danken: any idea who I could discuss about making a gluster firewalld service file? 20:15 < apuimedo> *about with 20:16 -!- phoracek (phoracek): has joined #vdsm 20:21 <@danken> apuimedo: recently, tjeyasin@redhat.com has been responsive 20:22 < apuimedo> cool 20:22 <@danken> I suggest that you simply open a BZ on him, and he'll redirect it. 20:22 < apuimedo> I'll remember that 20:22 < apuimedo> good idea 20:24 < apuimedo> going 20:24 < apuimedo> have a nice evening everybody 20:28 -!- rmatinata: has quit [Ping timeout: 250 seconds] 20:29 -!- apuimedo: has quit [Ping timeout: 258 seconds] 20:36 -!- rmatinata (Ricardo Marin Matinata): has joined #vdsm 20:42 -!- ybronhei (purple): has joined #vdsm 21:19 -!- #vdsm ybronhei: has quit [Quit: Leaving.] 21:31 -!- phoracek: has quit [Ping timeout: 272 seconds] 21:37 -!- adahms (Andrew Dahms): has joined #vdsm 22:14 -!- phoracek (phoracek): has joined #vdsm 22:16 -!- phoracek: has quit [Client Quit] 22:16 -!- phoracek (phoracek): has joined #vdsm 22:31 -!- adahms: has quit [Quit: Leaving] 22:33 -!- phoracek: has quit [Ping timeout: 260 seconds] 22:43 -!- #vdsm dougsland: has quit [Ping timeout: 272 seconds] 22:44 -!- #vdsm dyasny: has quit [Ping timeout: 250 seconds] 22:46 -!- rmatinata: has quit [Ping timeout: 250 seconds] 23:00 -!- phoracek (phoracek): has joined #vdsm 23:08 -!- dougsland (Douglas): has joined #vdsm 23:13 < nsoffer> danke, please take a look at http://gerrit.ovirt.org/33875 23:13 < nsoffer> danken, I would like to get it out in 3.5.0 GA if possible --- Log closed wo okt 08 00:00:54 2014