[01:16:46] *** Joins: tomzawadzki (tomzawadzk@nat/intel/x-ruzzivtudpwskjvl) [01:18:54] *** Quits: tomzawadzki (tomzawadzk@nat/intel/x-ruzzivtudpwskjvl) (Client Quit) [01:54:24] *** Quits: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) (Ping timeout: 260 seconds) [02:29:53] *** Joins: tomzawadzki (tomzawadzk@nat/intel/x-kocmdvytudzpfdyi) [03:35:55] *** Joins: tkulasek (86bfdc4a@gateway/web/freenode/ip.134.191.220.74) [05:55:55] *** Quits: tkulasek (86bfdc4a@gateway/web/freenode/ip.134.191.220.74) (Ping timeout: 260 seconds) [06:56:15] *** Quits: gila (~gila@5ED4D9C8.cm-7-5d.dynamic.ziggo.nl) (Ping timeout: 248 seconds) [07:01:17] *** Joins: gila (~gila@5ED4D9C8.cm-7-5d.dynamic.ziggo.nl) [09:37:05] *** Quits: tomzawadzki (tomzawadzk@nat/intel/x-kocmdvytudzpfdyi) (Ping timeout: 256 seconds) [13:54:52] drv, OK updated passthru example. I tried using bdevio before, as you suggested, and ran into a zillion problems. With the remove_cb fix from Jim and the latest code it seems to work fine now.... [13:55:25] PS: my patch includes Jim's fix since it's not merged yet and nothing works right without it [14:00:22] peluse: which review is Jim's fix? [14:00:32] the defer via message thing? [14:00:40] yeah [14:01:05] https://review.gerrithub.io/#/c/400305/ [14:01:16] I'll push it now - it has +2 from drv and shuhei already [14:01:20] I just pushed it [14:02:05] cool, so those changes will go away after I rebase but I'll wait for CI to be done with what I've got posted on https://review.gerrithub.io/#/c/399768/ [14:09:30] peluse: you may want to remove the "Entry IO type" noticelog too - it's currently generating ~20 MB log files :) [14:12:19] drv, yeah I will. That was just to make sure something was really happening during dev/test. I'll remove it when I rebase. Just look at the one failure remaining in CI on one of the Ubunutu machines, no logs? [14:13:09] the CI publish script won't copy any files larger than 2 MB - that's why it doesn't show up on ci.spdk.io [14:13:23] cool [14:13:27] the log is there on the internal spdk.intel.com machine [14:13:31] not sure why it failed, though - it just hit the timeout [14:13:54] maybe it just took too long with all the printouts? [14:15:28] *** Joins: lhodev (~Adium@inet-hqmc04-o.oracle.com) [14:17:36] guess we'll see.. [14:43:38] yeah, passed this time. Ready for review [15:45:36] drv: any reason we should not ignore SIGPIPE? [15:45:44] I think this is from the original istgt code [15:51:11] no idea [15:51:22] is that to do with running as a daemon? [15:51:22] *** Joins: Shuhei (caf6fc61@gateway/web/freenode/ip.202.246.252.97) [15:52:19] more if you're piping output of an SPDK process into another process - and that other process dies [15:52:32] yeah, but that could also be the controlling pty or something, right? [15:52:41] like if you background the process [15:56:13] I thought if the controlling shell exits that background processes get SIGHUP? [15:57:15] but yeah - i guess if controlling pty exits then stdout has nowhere to go [16:05:54] not sure [16:06:01] I'm OK with removing it [16:11:19] no - we need to keep it [16:11:26] now that i understand this more [16:12:09] for example, JSON RPC - if client sends us RPC message then exits without waiting for response, when we write response to socket, we would get SIGPIPE [16:12:27] ah, ok [16:23:49] *** Quits: lhodev (~Adium@inet-hqmc04-o.oracle.com) (Remote host closed the connection) [16:24:07] *** Joins: lhodev (~Adium@66-90-218-190.dyn.grandenetworks.net) [16:50:13] peluse: your community meeting blog post looks good now - I've pushed it [16:50:48] cool thanks [16:50:53] I'll send an email out [16:51:12] and for anyone reading this, go check out http://spdk.io for the latest on community meetings :) [16:59:28] I'll ping one of you guys tomorrow too to make sure the WebEx stuff is working correctly... [16:59:30] Hi All, I posted a question to mailing lists about mounting lvolstore. Why I asked was that using config file for lvolstore looked difficult because blobstore is persistent storage different from other libraries. [16:59:58] If this topic is good for meeting, I'll register it into Trello. Thanks. [17:01:08] Hi Paul, Daniel, I saw the new page, looks nice, thank you. [17:01:47] Shuhei, either way... you can add anything to Trello and if it gets covered before-hand we'll just skip it at the meeting. [17:03:08] Paul, OK, I'll register it anyway if my question is not totally wrong. thank you. [17:07:48] peluse: I didn't think of it before, but we should probably add some kind of calendar or other info to the community page so new folks can figure out which week it is in the cycle [17:08:20] the first one is fine since it's in the blog post, but it will start getting harder to figure out once we get farther along [17:09:54] yeah, will figure that out [17:10:47] i thought of a few way but they were more confusing than just sending a weekly note out but I'm sure we can some up with something easy to put on the website but I'm sick and not feeling all that creative today :) [17:11:53] yeah, no problem for now, just thought of that as I was putting a reminder on my calendar [17:12:52] we can probably add .ics (iCalendar) files so nobody has to do timezone math :) [17:13:07] I ain't doin' that [17:13:13] :) [23:24:00] *** Joins: tomzawadzki (~tomzawadz@134.134.139.76) [23:30:38] *** Quits: tomzawadzki (~tomzawadz@134.134.139.76) (Quit: Leaving) [23:30:49] *** Joins: tomzawadzki (~tomzawadz@134.134.139.76)