You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@trafficserver.apache.org by "Zhao Yongming (JIRA)" <ji...@apache.org> on 2014/11/08 15:20:33 UTC

[jira] [Created] (TS-3180) Linux native aio not support disk >2T

Zhao Yongming created TS-3180:
---------------------------------

             Summary: Linux native aio not support disk >2T
                 Key: TS-3180
                 URL: https://issues.apache.org/jira/browse/TS-3180
             Project: Traffic Server
          Issue Type: Bug
          Components: Core
            Reporter: Zhao Yongming


{code}
21:47 < faysal> [Nov  8 15:45:30.080] Server {0x2ab53ff36700} WARNING: unable to clear cache directory '/dev/sdc 548864:366283256'
21:48 < faysal> although brw-rw---- 1 nobody nobody 8, 32 Nov  8 15:45 /dev/sdc
21:48 < faysal> fedora 21
21:48 < faysal> ping anyone
21:49 < ming_zym> disk fail?
21:52 < ming_zym> try to restart traffic server?
21:55 < faysal> i did restarted traffic server couple of times no luck
21:56 < faysal> by the way this is build with linux native aio enabled
21:56 < faysal> and latest master pulled today
21:56 < ming_zym> o, please don't use linux native aio in production
21:57 < ming_zym> not that ready to be used expect in testing
21:58 < ming_zym> I am sorry we don't have time to track down all those native aio issues here
21:59 < faysal> ok
21:59 < faysal> am compiling now without native aio
21:59 < faysal> and see what happens and inform you
22:06 < faysal> ming_zym: if you are working on native aio stuff its the issue
22:07 < faysal> i compiled without it and now its working fine
22:07 < faysal> i have noticed this on harddisks over 2T size
22:07 < faysal> smaller disks work fine with native aio
22:12 < ming_zym> ok, cool
22:13 < faysal> thats because i guess my disks are 3T each and one with 240G
22:14 < faysal> the 240 was taken no problem
22:14 < faysal> but the 3T has to be in GPT patition format
22:14 < faysal> and Fedora for some reason had issues identifying it
22:14 < ming_zym> hmm, maybe that is bug
{code}



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)