| *** Masahiro has joined #openstack-swift | 00:07 | |
| *** Masahiro has quit IRC | 00:12 | |
| *** dmorita has joined #openstack-swift | 00:29 | |
| *** bill_az has quit IRC | 00:44 | |
| dmorita | Hi, guys. Could someone give feedback to https://review.openstack.org/#/c/123765/ | 00:51 |
|---|---|---|
| dmorita | This patch remains unreviewed for two months. | 00:52 |
| *** nomad411 has joined #openstack-swift | 00:57 | |
| *** Masahiro has joined #openstack-swift | 01:02 | |
| *** addnull has joined #openstack-swift | 01:04 | |
| *** ekarlso- has quit IRC | 01:12 | |
| *** X019 has joined #openstack-swift | 01:12 | |
| *** ekarlso- has joined #openstack-swift | 01:13 | |
| *** addnull has quit IRC | 01:27 | |
| *** rdaly2 has joined #openstack-swift | 01:29 | |
| *** rdaly2 has quit IRC | 01:33 | |
| *** rmcall has quit IRC | 01:34 | |
| *** nomad411 has quit IRC | 01:35 | |
| *** abhirc has quit IRC | 01:36 | |
| *** NM has joined #openstack-swift | 01:37 | |
| *** haomaiwang has quit IRC | 01:42 | |
| *** panbalag has quit IRC | 01:42 | |
| *** panbalag has joined #openstack-swift | 01:45 | |
| *** Masahiro has quit IRC | 01:56 | |
| *** NM has quit IRC | 01:57 | |
| *** haomaiwang has joined #openstack-swift | 02:02 | |
| *** tellesnobrega_ has quit IRC | 02:04 | |
| *** addnull has joined #openstack-swift | 02:06 | |
| *** Masahiro has joined #openstack-swift | 02:10 | |
| *** nosnos has joined #openstack-swift | 02:11 | |
| *** fifieldt has joined #openstack-swift | 02:13 | |
| *** tellesnobrega_ has joined #openstack-swift | 02:18 | |
| *** addnull has quit IRC | 02:18 | |
| *** addnull has joined #openstack-swift | 02:20 | |
| *** tellesnobrega_ has quit IRC | 02:29 | |
| *** mahatic has quit IRC | 02:34 | |
| *** rdaly2 has joined #openstack-swift | 02:41 | |
| *** tellesnobrega_ has joined #openstack-swift | 02:44 | |
| *** rdaly2 has quit IRC | 02:45 | |
| *** addnull has quit IRC | 03:00 | |
| *** Masahiro has quit IRC | 03:06 | |
| *** addnull has joined #openstack-swift | 03:22 | |
| *** bkopilov has quit IRC | 03:22 | |
| *** X019 has quit IRC | 03:38 | |
| clayg | notmyname: if you ment to +2 https://review.openstack.org/#/c/133315/ will you go back and merge it - sam said he tried a native check_utf8 and it stupid slow | 03:48 |
| clayg | notmyname: i'm not really worried about the few extra string comparisions in the grand scheme of things, just mad at python | 03:48 |
| clayg | notmyname: torgomatic: OH! so we can collapse them both into one - sam's check for surrogates would also return False on the round trip bug - duh - just do them in the other order and the decode.encode thing is noop | 03:52 |
| *** Masahiro has joined #openstack-swift | 03:54 | |
| clayg | oh my god | 03:55 |
| * clayg face palms | 03:55 | |
| clayg | notmyname: go ahead and merge that change when you get a chance pls | 03:56 |
| clayg | i'm apparently red/green color blind today | 03:56 |
| *** nosnos has quit IRC | 04:00 | |
| *** anderstj_ has joined #openstack-swift | 04:05 | |
| *** anderstj has quit IRC | 04:09 | |
| *** anderstj_ is now known as anderstj | 04:09 | |
| *** tellesnobrega_ has quit IRC | 04:27 | |
| *** oomichi_ has joined #openstack-swift | 04:30 | |
| *** Masahiro has quit IRC | 04:41 | |
| *** Masahiro has joined #openstack-swift | 04:45 | |
| *** nosnos has joined #openstack-swift | 04:49 | |
| *** addnull has quit IRC | 04:52 | |
| *** rebelshrug has quit IRC | 04:52 | |
| *** ppai has joined #openstack-swift | 04:57 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:05 | |
| *** SkyRocknRoll has joined #openstack-swift | 05:05 | |
| *** bkopilov has joined #openstack-swift | 05:06 | |
| *** kopparam has joined #openstack-swift | 05:13 | |
| *** addnull has joined #openstack-swift | 05:21 | |
| *** addnull has quit IRC | 05:21 | |
| *** addnull has joined #openstack-swift | 05:21 | |
| *** Masahiro has quit IRC | 05:22 | |
| *** kopparam has quit IRC | 05:23 | |
| *** addnull has quit IRC | 05:23 | |
| *** kopparam has joined #openstack-swift | 05:26 | |
| *** addnull has joined #openstack-swift | 05:26 | |
| *** kopparam has quit IRC | 05:29 | |
| *** kopparam has joined #openstack-swift | 05:29 | |
| *** Masahiro has joined #openstack-swift | 05:32 | |
| *** kopparam has quit IRC | 05:33 | |
| *** kopparam has joined #openstack-swift | 05:33 | |
| *** X019 has joined #openstack-swift | 05:34 | |
| *** panbalag has quit IRC | 05:37 | |
| *** atan8 has joined #openstack-swift | 05:38 | |
| *** addnull has quit IRC | 05:40 | |
| *** addnull has joined #openstack-swift | 05:40 | |
| *** addnull_ has joined #openstack-swift | 05:44 | |
| *** addnull has quit IRC | 05:45 | |
| *** Masahiro has quit IRC | 06:04 | |
| *** panbalag has joined #openstack-swift | 06:05 | |
| *** Masahiro has joined #openstack-swift | 06:08 | |
| *** sungju has quit IRC | 06:10 | |
| *** sungju has joined #openstack-swift | 06:12 | |
| *** echevemaster has quit IRC | 06:13 | |
| *** gvernik has joined #openstack-swift | 06:21 | |
| *** k4n0 has joined #openstack-swift | 06:26 | |
| *** ppai has quit IRC | 06:34 | |
| *** sungju has quit IRC | 06:42 | |
| *** panbalag has quit IRC | 06:46 | |
| *** ppai has joined #openstack-swift | 06:48 | |
| *** ekarlso- has quit IRC | 06:54 | |
| *** ekarlso- has joined #openstack-swift | 06:54 | |
| *** zaitcev has quit IRC | 06:59 | |
| *** kopparam has quit IRC | 07:04 | |
| *** Masahiro has quit IRC | 07:16 | |
| *** Masahiro has joined #openstack-swift | 07:19 | |
| *** nshaikh has joined #openstack-swift | 07:21 | |
| *** fed007 has joined #openstack-swift | 07:25 | |
| fed007 | hi | 07:26 |
| fed007 | Can anyone help with Could not autocreate account '/AUTH_test' | 07:28 |
| fed007 | doing some SAIO tests and this is the first time I have run into this | 07:29 |
| *** ppai has quit IRC | 07:41 | |
| *** kopparam has joined #openstack-swift | 07:44 | |
| cschwede | fed007: permissions are set correctly on the swift storage directories? and account_autocreate is set to true in the swift proxy config? | 07:52 |
| *** Trixboxer has joined #openstack-swift | 07:53 | |
| *** ppai has joined #openstack-swift | 07:54 | |
| fed007 | yes all is root in these tests | 08:01 |
| fed007 | I can do a few pastes | 08:02 |
| fed007 | http://ur1.ca/iwm56 | 08:04 |
| fed007 | remakering http://ur1.ca/iwm5t | 08:06 |
| fed007 | to isolate out my Swiftonfile tests I went back to a basic config | 08:06 |
| fed007 | which is still crashing at the autocreate account | 08:07 |
| fed007 | everything is root so perms should be good.... a ps -aux shows all PID running as root.... | 08:07 |
| mattoliverau | fed007: your using fpaste, does this mean your using fedora/centos/redhat if so is selinux clobbering anything? | 08:11 |
| fed007 | no this is ubunt 12.04 | 08:12 |
| mattoliverau | OK :( | 08:12 |
| fed007 | I'll make sure se is disabled | 08:12 |
| fed007 | if it is there... | 08:13 |
| fed007 | so if I run remakering it should create a db file with the tempauth users right? | 08:13 |
| fed007 | can I use the swift-account-info command to verify anything or do we still have a write issue | 08:14 |
| fed007 | If I look in the /srv/node/pod1,2,3,4 folders all is empty | 08:16 |
| fed007 | there should be accounts folders and container folders and the hash name part folders... | 08:17 |
| mattoliverau | Remake rings just makes the rings, you need to have auto create accounts turned on for the temp auth db to be created | 08:21 |
| mattoliverau | Sorry on phone so autpcomplete is clobbering things | 08:21 |
| fed007 | np | 08:22 |
| fed007 | my proxy file... http://ur1.ca/iwmdu autocreate set to true | 08:23 |
| cschwede | fed007: are you using vagrant+virtualbox by chance? | 08:24 |
| fed007 | should some process be crating a passwd file? | 08:24 |
| *** oomichi_ has quit IRC | 08:24 | |
| *** fed007 has quit IRC | 08:25 | |
| *** fed007 has joined #openstack-swift | 08:25 | |
| fed007 | syslog when trying to PUT a new container on the default test account http://ur1.ca/iwmgx | 08:30 |
| *** Masahiro has quit IRC | 08:30 | |
| *** addnull_ has quit IRC | 08:30 | |
| cschwede | fed007: all services are running? especially account/container/object servers? | 08:32 |
| *** aswadr has joined #openstack-swift | 08:32 | |
| *** Masahiro has joined #openstack-swift | 08:33 | |
| fed007 | yeah I am looking at ps and there seem to be double the amount I would expect ATM let me killall and restart | 08:34 |
| mattoliverau | Hi cschwede! | 08:35 |
| cschwede | Hello mattoliverau! | 08:35 |
| fed007 | look at this... http://fpaste.org/154575/70773631/raw/ | 08:36 |
| cschwede | fed007: looking at http://fpaste.org/154572/17075538/ - are you sure your ports are correct set in the ring? and match your config? | 08:36 |
| fed007 | should there only be one for each type? | 08:37 |
| fed007 | account server has double processes.... is that normal? | 08:38 |
| mattoliverau | Well I'm about to board my plane home, the long flying day begins :( good luck fed007, your in good hands. I'll do what I can to help if your still having problems when I get home (in about 25 - 30 hours) but will probably need sleep then :p | 08:38 |
| fed007 | lol | 08:38 |
| fed007 | bon voyage | 08:38 |
| fed007 | yes my ports are correct ... I am explicity using 6001 -6004 for my base object server | 08:39 |
| fed007 | the rest match the SAIO tut | 08:39 |
| cschwede | mattoliverau: travelling to au now? | 08:40 |
| cschwede | mattoliverau: have a safe trip back! | 08:41 |
| ahale | looks like those processes are main processes and workers , or whatever the term is.. does it look more normal with a tree view ? | 08:41 |
| mattoliverau | cschwede: yeah, my 3 week holiday is up, back at work on Monday. Leaving Paris now :) thanks enjoy the rest of you week! | 08:41 |
| fed007 | yep those are work threads - false alarm there | 08:42 |
| cschwede | mattoliverau: i have been so busy the last days, already forgot that the summit is only 2,5 ago… anyways, safe trip, i hope you get some rest! see you soon | 08:43 |
| cschwede | s/2,5/2,5 weeks/ | 08:44 |
| cschwede | fed007: hmm, strange. can you paste the output from „netstat -atnlp“? | 08:45 |
| *** ppai has quit IRC | 08:47 | |
| fed007 | http://fpaste.org/154580/14170780/raw/ | 08:48 |
| *** nosnos has quit IRC | 08:49 | |
| fed007 | is there anything we should be trying bind 0.0.0.0 on? | 08:49 |
| *** mmcardle has joined #openstack-swift | 08:49 | |
| cschwede | fed007: not if everything runs only local. do you see any other logmessages apart from the proxy? | 08:51 |
| *** jordanP has joined #openstack-swift | 08:51 | |
| fed007 | I'll check... no python traces for sure | 08:52 |
| fed007 | http://fpaste.org/154581/17078281/raw/ here is my /etc/swift any files missing from here? | 08:52 |
| *** kopparam has quit IRC | 08:52 | |
| fed007 | I think we should be able to verify that the ringbuilder built the account test into somewhere | 08:53 |
| fed007 | (where I am not sure) | 08:53 |
| *** kopparam has joined #openstack-swift | 08:56 | |
| fed007 | http://fpaste.org/154583/41707870/raw/ | 08:58 |
| fed007 | here is some interesting part : Invalid path: /srv/node/pod2/3310/AUTH_test | 08:59 |
| fed007 | on the GET | 08:59 |
| fed007 | the PUT dies on the autocreate | 08:59 |
| fed007 | it is almost as though another db file is being referenced , where else could db file reside? | 09:00 |
| *** ppai has joined #openstack-swift | 09:00 | |
| fed007 | or is this type of path data in specific ring file? | 09:01 |
| *** mmcardle has quit IRC | 09:05 | |
| *** foexle has joined #openstack-swift | 09:07 | |
| *** [1]gvernik has joined #openstack-swift | 09:09 | |
| *** jistr has joined #openstack-swift | 09:10 | |
| *** gvernik has quit IRC | 09:11 | |
| *** [1]gvernik is now known as gvernik | 09:11 | |
| fed007 | does /srv/node/ need to contain a mount point? I switched this to just be a dir path | 09:12 |
| fed007 | on the root fs | 09:12 |
| fed007 | mkdir /srv/node/pod1 ..... | 09:12 |
| *** addnull has joined #openstack-swift | 09:13 | |
| *** nellysmitt has joined #openstack-swift | 09:14 | |
| *** Masahiro has quit IRC | 09:16 | |
| *** Masahiro has joined #openstack-swift | 09:21 | |
| ipolyzos | notmyname: good morning | 09:24 |
| ipolyzos | notmyname: thanks for the snippet yesterday | 09:24 |
| ipolyzos | the user id you are using is user_name:tenant:name ? | 09:24 |
| *** mmcardle has joined #openstack-swift | 09:25 | |
| ahale | fed007: it should be alright just as a dir in a saio, maybe check for the check mount conf flag - and you can check what paths etc. are from the rings with swift-get-nodes | 09:28 |
| fed007 | ok here is the output of that check http://fpaste.org/154590/17080608/raw/ | 09:30 |
| fed007 | should I manually create thos paths? | 09:32 |
| ahale | nah, it should be fine - the stuff after pod4 in '/srv/node/pod4/accounts/54/f6c/' will be created by swift, as long as it has write etc | 09:32 |
| ahale | off that the invalid path error earlier is different to the paths from ring.. not sure whats up there | 09:33 |
| fed007 | should I clear a cache somewhere? | 09:34 |
| fed007 | memcache /var cache...? | 09:34 |
| ahale | i would probably end up stracing the account servers, seeing what they are actually trying to do and what call fails | 09:34 |
| ahale | restarting the memcache could help if it has some dodgy account info in it maybe | 09:34 |
| fed007 | Do you think this is some cache issue ? | 09:35 |
| fed007 | I could add a new user in the temp auth section | 09:35 |
| fed007 | see if that one will not be 'corrupt' somehow | 09:35 |
| ahale | i dont think its likely, but really not too sure what it would be | 09:35 |
| fed007 | well this used to have a swiftstack deploy on it | 09:36 |
| fed007 | but I did an apt-get purge remove everything,... | 09:37 |
| fed007 | then dropped out the ssnode... | 09:37 |
| fed007 | killed all the folders in opt... | 09:37 |
| ahale | ah havent played with their build so dont really know if that could be relevant | 09:37 |
| fed007 | I dont see any relevenat env vars... | 09:38 |
| fed007 | searches return singletons of the configs... | 09:38 |
| *** addnull has quit IRC | 09:38 | |
| *** nosnos has joined #openstack-swift | 09:39 | |
| ahale | if you have the resetswift script - the stuff in there shows everythingh thats needed to remove traces of existing stuff | 09:39 |
| ahale | i think its just 'delete everything from disks and restart memcache' though | 09:40 |
| fed007 | ok I am trying a memcache restart then a new user... What is wiered to me is that a fresh set of GIT files should be the ones run - unsless some python packages from swiftstack are buried in some core python paths | 09:42 |
| fed007 | but they apparently run core swift .... | 09:42 |
| fed007 | (and I ran their uninstaller) | 09:44 |
| ahale | yah, they know what they are doing so i would expect their uninstaller to be decent | 09:46 |
| ahale | they'll maybe be around later , west coast states but then thanksgiving today so who knows | 09:46 |
| *** rdaly2 has joined #openstack-swift | 09:46 | |
| fed007 | same deal with memcache restart .... why does the ring db think the user account should be at a path ? or did it assign but not able to create... | 09:47 |
| fed007 | I'll try the user... if not it MUST be a write issue | 09:48 |
| ahale | the ring is just a lookup for where stuff is stored, the user account is just a sqlite db in a dir defined by rings mapings of hash to partitions | 09:48 |
| fed007 | what level is the sqlite db stored at - fairly high in the dir structure? | 09:49 |
| *** rdaly2 has quit IRC | 09:51 | |
| ahale | yeah, it would be inside the /srv/node/pod4/accounts/54/f6c/036599eec10cb71a38ab9a43a9c7cf6c directory | 09:51 |
| ahale | like: -rw------- 1 swift swift 57K 2014-11-27 09:25 /srv/node/c0u1/accounts/41061/af8/281962d939d59a3e9df92adea5288af8/281962d939d59a3e9df92adea5288af8.db | 09:52 |
| fed007 | is there a default user/pass to open the sqlite? | 09:54 |
| fed007 | or just go wild | 09:55 |
| ahale | just go wild :) | 09:55 |
| fed007 | http://ur1.ca/iwnit (output from new user ) | 09:55 |
| fed007 | So it must be assigning the /srv/node/pod4/4056/AUTH_fed007 in memory for output messaging but failing on the write | 09:56 |
| fed007 | because it is not existing ... | 09:56 |
| ahale | yeah i dunno what it is doing there at all - that invalid path is weird | 09:57 |
| fed007 | so I pretty much chmodded them all to 777 and still no write | 09:57 |
| fed007 | how many base.py do you have in your sys? | 09:59 |
| fed007 | http://fpaste.org/154599/82318141/raw/ | 09:59 |
| ahale | lemme look | 10:00 |
| fed007 | is there a pkg / module dir I can turf on my end? | 10:00 |
| *** [1]gvernik has joined #openstack-swift | 10:00 | |
| ahale | i just have '/usr/lib/pymodules/python2.6/swift/proxy/controllers/base.py' and '/usr/share/pyshared/swift/proxy/controllers/base.py' | 10:01 |
| ahale | which looks pretty similar, since i dont have source locally | 10:02 |
| ahale | might be best asking around later if anyone else has ideas - im all out | 10:02 |
| fed007 | yeah cloning from github and building makes for my extras | 10:02 |
| *** gvernik has quit IRC | 10:03 | |
| fed007 | np I did learn form you :) | 10:03 |
| *** [1]gvernik is now known as gvernik | 10:03 | |
| fed007 | thank-you for your help | 10:03 |
| ahale | no problem, sorry it wasnt more helpful | 10:03 |
| fed007 | lol - well we are closer... | 10:03 |
| fed007 | I want this box testing the swiftonfile and do not want to do a reinstall just for swift | 10:04 |
| fed007 | the LSI HBA card for the 24 drive bay was minor hell to get configure | 10:04 |
| fed007 | now that it runs Iam just wanting to play with it more and the new swiftonfile policy which is totally aweome in another glusterfs test I have working | 10:05 |
| ahale | haha i know your pain - looks at his percs and jbods | 10:05 |
| fed007 | hehe - yeah a few python packages and software config I'll take ;) | 10:06 |
| fed007 | OK well I am on 3 am for day 3 in a row | 10:06 |
| fed007 | I feel like goin more at this unit but I am out of ideas at the momemnt too | 10:07 |
| ahale | ouch | 10:07 |
| fed007 | It's all good I just slepp till noon lol | 10:08 |
| fed007 | OK I have a question... | 10:09 |
| fed007 | are you sure about the dir path for the base of /srv/node? | 10:09 |
| *** atan8 has quit IRC | 10:09 | |
| fed007 | there is a lot of switches about mount checks | 10:09 |
| fed007 | maybe it looks for a vol id or some attribute to ensure device diff | 10:10 |
| ahale | thats the default so unless you specified it differently in one of the confs it should be | 10:10 |
| fed007 | yeah I agree, it is implied that it is just paths and it should matter if they go through mount points or not | 10:11 |
| openstackgerrit | Daisuke Morita proposed openstack/swift: Output logs of policy index https://review.openstack.org/136995 | 10:11 |
| *** geaaru has joined #openstack-swift | 10:11 | |
| fed007 | err - shouldn't | 10:11 |
| *** sungju has joined #openstack-swift | 10:12 | |
| *** sungju has quit IRC | 10:12 | |
| ahale | yah it should just be paths too, it shoudn't worry about labels or ids | 10:12 |
| ahale | we see that when a disk gets accidently remounted on the wrong point, its not treated as the old disk.. just as one with a load of out of place data | 10:13 |
| *** Masahiro has quit IRC | 10:13 | |
| fed007 | I might have find the code where is tries to make those directories | 10:14 |
| *** acoles_away is now known as acoles | 10:14 | |
| *** atan8 has joined #openstack-swift | 10:14 | |
| fed007 | https://github.com/openstack/swift/blob/master/swift/proxy/controllers/base.py | 10:14 |
| fed007 | this one has the errorr message when I try to create a container... I get a 404 and the "unable to autocreate" | 10:15 |
| *** tellesnobrega_ has joined #openstack-swift | 10:15 | |
| *** Masahiro_ has joined #openstack-swift | 10:20 | |
| *** Masahiro_ has quit IRC | 10:22 | |
| fed007 | alright I'll be back to solve again | 10:25 |
| fed007 | goodnight! | 10:25 |
| *** fed007 has quit IRC | 10:29 | |
| *** tellesnobrega_ has quit IRC | 10:31 | |
| *** Masahiro has joined #openstack-swift | 10:32 | |
| *** Masahiro has quit IRC | 10:34 | |
| *** tellesnobrega_ has joined #openstack-swift | 10:35 | |
| *** kopparam has quit IRC | 10:36 | |
| *** kopparam has joined #openstack-swift | 10:37 | |
| *** ppai has quit IRC | 10:37 | |
| *** aix has joined #openstack-swift | 10:37 | |
| *** Masahiro has joined #openstack-swift | 10:37 | |
| openstackgerrit | Daniel Wakefield proposed openstack/python-swiftclient: Verify MD5 of uploaded objects. https://review.openstack.org/129254 | 10:48 |
| *** addnull has joined #openstack-swift | 10:49 | |
| *** addnull has joined #openstack-swift | 10:50 | |
| *** ppai has joined #openstack-swift | 10:50 | |
| *** kopparam has quit IRC | 10:52 | |
| *** kopparam_ has joined #openstack-swift | 10:52 | |
| *** tellesnobrega_ has quit IRC | 10:52 | |
| *** NM has joined #openstack-swift | 10:52 | |
| *** Masahiro has quit IRC | 10:58 | |
| *** fifieldt has quit IRC | 10:59 | |
| *** kopparam_ has quit IRC | 11:01 | |
| *** addnull has quit IRC | 11:02 | |
| *** gvernik has quit IRC | 11:02 | |
| *** mandarine has joined #openstack-swift | 11:04 | |
| mandarine | Good morning | 11:04 |
| openstackgerrit | Merged openstack/swift: Fix tempfile bugs from commit 6978275 https://review.openstack.org/137482 | 11:06 |
| openstackgerrit | Daisuke Morita proposed openstack/swift: Output logs of policy index https://review.openstack.org/136995 | 11:14 |
| *** kopparam has joined #openstack-swift | 11:15 | |
| *** haomaiwang has quit IRC | 11:27 | |
| *** NM has quit IRC | 11:30 | |
| *** NM has joined #openstack-swift | 11:30 | |
| *** haomaiwang has joined #openstack-swift | 11:44 | |
| *** rdaly2 has joined #openstack-swift | 11:48 | |
| *** rdaly2 has quit IRC | 11:52 | |
| *** aix has quit IRC | 11:55 | |
| *** kopparam has quit IRC | 11:55 | |
| *** kopparam has joined #openstack-swift | 11:58 | |
| *** exploreshaifali has joined #openstack-swift | 12:02 | |
| *** aswadr has quit IRC | 12:10 | |
| *** ppai has quit IRC | 12:11 | |
| *** Masahiro has joined #openstack-swift | 12:12 | |
| *** addnull has joined #openstack-swift | 12:13 | |
| *** addnull has quit IRC | 12:13 | |
| *** addnull has joined #openstack-swift | 12:14 | |
| *** Masahiro has quit IRC | 12:15 | |
| *** Masahiro has joined #openstack-swift | 12:16 | |
| *** Masahiro has quit IRC | 12:18 | |
| *** Masahiro has joined #openstack-swift | 12:25 | |
| *** ppai has joined #openstack-swift | 12:25 | |
| *** k4n0 has quit IRC | 12:26 | |
| *** ppai has quit IRC | 12:26 | |
| *** Masahiro has quit IRC | 12:29 | |
| *** dmorita has quit IRC | 12:29 | |
| *** mmcardle has quit IRC | 12:44 | |
| *** mahatic has joined #openstack-swift | 12:49 | |
| *** panbalag has joined #openstack-swift | 12:55 | |
| *** dencaval has joined #openstack-swift | 13:01 | |
| *** nosnos has quit IRC | 13:01 | |
| *** exploreshaifali has quit IRC | 13:03 | |
| *** aix has joined #openstack-swift | 13:06 | |
| *** atan8 has quit IRC | 13:11 | |
| *** kopparam has quit IRC | 13:13 | |
| *** leopoldj has joined #openstack-swift | 13:20 | |
| *** mmcardle has joined #openstack-swift | 13:23 | |
| eikke | swifterdarrell: *poke* ;-) | 13:35 |
| *** rdaly2 has joined #openstack-swift | 13:49 | |
| mandarine | It seems I ran into a strange permission behaviour on swift partitions on one of my prod servers | 13:53 |
| mandarine | Some of the directory have a mode set to 0700 and some other are 0766 | 13:54 |
| mandarine | directories* | 13:54 |
| *** rdaly2 has quit IRC | 13:54 | |
| mandarine | The strange part is that the directories with permissions set to 0766 is only on one device ... | 13:55 |
| mandarine | I'm trying to investigate what happened, but if someone encontered this before, I'm open to suggestions | 13:55 |
| *** SkyRocknRoll has quit IRC | 13:57 | |
| mandarine | My bad, I wrote 766 when it is 755 | 14:00 |
| *** addnull has quit IRC | 14:00 | |
| mandarine | Still, I'm not sure what happened | 14:00 |
| ahale | some made by swift and some by rsync ? | 14:01 |
| *** addnull has joined #openstack-swift | 14:01 | |
| mandarine | Oh, maybe. I forgot about the rsync guy | 14:01 |
| mandarine | ahale: thank you very much. rsync seems to be the final source of my problem | 14:02 |
| ahale | haha yeah rsync is the source of many problems :) no worries | 14:02 |
| *** bkopilov has quit IRC | 14:02 | |
| mandarine | ahale: it make TONS of I/O on a single device | 14:03 |
| mandarine | I think I forgot something | 14:04 |
| ahale | maybe, but swift can be extreme too | 14:06 |
| *** Masahiro has joined #openstack-swift | 14:06 | |
| ahale | like- crazy high load, or cpu or io.. isn't a rare thing in my prod at times.. adding boxes, filling up drives, controller/disk failures etc. all have noticable impacts | 14:08 |
| mandarine | Yeah, but in my case it happens on a single drive | 14:10 |
| mandarine | Specifically the first drive of the first server | 14:10 |
| *** Masahiro has quit IRC | 14:11 | |
| *** addnull has quit IRC | 14:13 | |
| *** NM1 has joined #openstack-swift | 14:29 | |
| *** NM has quit IRC | 14:31 | |
| *** infotection has quit IRC | 14:32 | |
| *** theanalyst has quit IRC | 14:33 | |
| *** infotection has joined #openstack-swift | 14:33 | |
| *** theanalyst has joined #openstack-swift | 14:38 | |
| *** sandywalsh has quit IRC | 14:40 | |
| *** acoles is now known as acoles_away | 14:40 | |
| *** kopparam has joined #openstack-swift | 14:49 | |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift: Support of the Linux socket option SO_REUSEPORT https://review.openstack.org/137659 | 15:03 |
| *** foexle has quit IRC | 15:03 | |
| *** NM1 has quit IRC | 15:28 | |
| *** NM has joined #openstack-swift | 15:33 | |
| *** rdaly2 has joined #openstack-swift | 15:51 | |
| *** rdaly2 has quit IRC | 15:55 | |
| *** Masahiro has joined #openstack-swift | 15:55 | |
| *** kopparam has quit IRC | 15:59 | |
| *** Masahiro has quit IRC | 16:00 | |
| *** nshaikh has quit IRC | 16:16 | |
| *** rledisez has joined #openstack-swift | 16:29 | |
| *** kopparam has joined #openstack-swift | 16:30 | |
| *** mahatic has quit IRC | 16:36 | |
| *** bkopilov has joined #openstack-swift | 16:38 | |
| *** nellysmitt has quit IRC | 16:46 | |
| *** mahatic has joined #openstack-swift | 16:54 | |
| NM | Hey guys! | 17:02 |
| NM | Is there any difference between create an all new ring when adding a server instead of adding a server to the existing ring ? | 17:03 |
| NM | Besides the extra work… | 17:03 |
| ahale | a new ring would be bad, you'd shuffle all the partitions and have to do a ton of replication to recover while modifying the existing will minimise partition movements and impact | 17:10 |
| openstackgerrit | Romain LE DISEZ proposed openstack/swift: Support of the Linux socket option SO_REUSEPORT https://review.openstack.org/137659 | 17:14 |
| *** kopparam has quit IRC | 17:14 | |
| NM | ahale: Thank you. We're changing the number of replicas here. | 17:18 |
| NM | I've just found the 'set_replicas' option. But the first plan was to create a new ring. | 17:19 |
| *** rledisez has quit IRC | 17:21 | |
| NM | ahale: Do you think it could cause loss of data? | 17:21 |
| *** vintbra has joined #openstack-swift | 17:25 | |
| *** bkopilov has quit IRC | 17:30 | |
| *** nellysmitt has joined #openstack-swift | 17:31 | |
| *** Masahiro has joined #openstack-swift | 17:44 | |
| *** exploreshaifali has joined #openstack-swift | 17:48 | |
| *** Masahiro has quit IRC | 17:48 | |
| *** jistr has quit IRC | 17:50 | |
| *** rdaly2 has joined #openstack-swift | 17:52 | |
| *** rdaly2 has quit IRC | 17:57 | |
| *** foexle has joined #openstack-swift | 18:11 | |
| ahale | im not sure NM i havent played with changing number of replicas at all | 18:12 |
| *** jordanP has quit IRC | 18:12 | |
| NM | ahale: I'll take the safest way. Thank you so much. | 18:15 |
| *** atan8 has joined #openstack-swift | 18:15 | |
| *** kopparam has joined #openstack-swift | 18:19 | |
| *** vintbra has quit IRC | 18:19 | |
| *** vintbra has joined #openstack-swift | 18:24 | |
| *** kopparam has quit IRC | 18:24 | |
| *** vintbra1 has joined #openstack-swift | 18:25 | |
| *** vintbra has quit IRC | 18:29 | |
| *** leopoldj has quit IRC | 18:39 | |
| *** foexle has quit IRC | 18:46 | |
| *** sandywalsh has joined #openstack-swift | 18:51 | |
| *** geaaru has quit IRC | 18:53 | |
| *** X019 has quit IRC | 19:11 | |
| *** aix has quit IRC | 19:15 | |
| *** X019 has joined #openstack-swift | 19:24 | |
| *** Masahiro has joined #openstack-swift | 19:33 | |
| *** Masahiro has quit IRC | 19:37 | |
| openstackgerrit | Merged openstack/swift: Reject object names with Unicode surrogates https://review.openstack.org/133315 | 19:40 |
| *** nellysmitt has quit IRC | 20:02 | |
| *** mahatic has quit IRC | 20:05 | |
| *** wer has quit IRC | 20:08 | |
| *** wer has joined #openstack-swift | 21:05 | |
| *** nomad411 has joined #openstack-swift | 21:10 | |
| *** NM has quit IRC | 21:12 | |
| *** Masahiro has joined #openstack-swift | 21:21 | |
| *** Masahiro has quit IRC | 21:26 | |
| openstackgerrit | OpenStack Proposal Bot proposed openstack/swift: Updated from global requirements https://review.openstack.org/88736 | 21:49 |
| *** kopparam has joined #openstack-swift | 21:57 | |
| *** kopparam has quit IRC | 22:02 | |
| *** nellysmitt has joined #openstack-swift | 22:03 | |
| *** nellysmitt has quit IRC | 22:07 | |
| *** vintbra1 has quit IRC | 22:26 | |
| *** vintbra has joined #openstack-swift | 22:34 | |
| *** vintbra1 has joined #openstack-swift | 22:38 | |
| *** vintbra has quit IRC | 22:38 | |
| *** vintbra1 has quit IRC | 22:40 | |
| *** atan8 has quit IRC | 22:42 | |
| *** atan8 has joined #openstack-swift | 22:45 | |
| *** atan8 has quit IRC | 22:58 | |
| *** kopparam has joined #openstack-swift | 23:03 | |
| *** kopparam has quit IRC | 23:09 | |
| *** Masahiro has joined #openstack-swift | 23:10 | |
| *** Masahiro has quit IRC | 23:15 | |
| *** tellesnobrega_ has joined #openstack-swift | 23:21 | |
| *** X019 has quit IRC | 23:26 | |
| *** oomichi has joined #openstack-swift | 23:43 | |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!