Sunday, 2018-09-16

timburkeit's a little discouraging when you're trying to figure out why we're masking errno=0 "errors" and you find https://github.com/openstack/swift/blame/e630e7c/swift/common/utils.py#L160 ...00:14
timburkemaybe that's because of the latent bug i noticed when dealing with posix_fallocate?00:14
*** rcernin has joined #openstack-swift00:31
*** early` has quit IRC03:05
*** early has joined #openstack-swift03:18
*** e0ne has joined #openstack-swift04:10
*** e0ne_ has joined #openstack-swift04:13
*** e0ne has quit IRC04:16
*** e0ne_ has quit IRC05:30
*** e0ne has joined #openstack-swift05:39
alecuyertimburke: I looked at this on the plane, I think you're right. It would also mean 2ca379d40f06d892fdf6e3430ccd8a78726e5626 would not work as intended06:08
alecuyerI think we would to do something like, if ret not in (ENOSYS, … , EINVAL): raise .. for posix_fallocate.  and for fallocate, remove 0 and EINVAL from the ignored error list06:09
alecuyer(haven't checked the relevant error codes yet for posix_fallocate beyond EINVAL)06:09
alecuyeroops the posix_fallocate would be ret not int (0, <errors we want to ignore>). Will look again after getting some sleep :)06:11
*** e0ne has quit IRC06:46
*** rcernin has quit IRC08:17
*** fatema__ has joined #openstack-swift09:50
*** fatema__ has quit IRC10:21
*** silor has joined #openstack-swift10:37
*** silor has quit IRC12:17
*** two_tired2 has joined #openstack-swift14:06
*** e0ne has joined #openstack-swift15:33
*** e0ne has quit IRC15:35
*** SkyRocknRoll has joined #openstack-swift16:42
*** SkyRocknRoll has quit IRC17:25
*** pcaruana has joined #openstack-swift18:48
*** pcaruana has quit IRC19:07
*** emccormickva has joined #openstack-swift19:22
*** emccormickva has quit IRC19:22
*** jistr has quit IRC20:47
*** jistr has joined #openstack-swift20:53
*** two_tired2 has quit IRC21:39
*** rcernin has joined #openstack-swift22:44

Generated by irclog2html.py 2.15.3 by Marius Gedminas - find it at mg.pov.lt!