You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@trafficserver.apache.org by "James Peach (JIRA)" <ji...@apache.org> on 2016/08/26 19:06:20 UTC

[jira] [Commented] (TS-4790) stale_while_revalidate crashing on task threads.

    [ https://issues.apache.org/jira/browse/TS-4790?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15439595#comment-15439595 ] 

James Peach commented on TS-4790:
---------------------------------

Every time the plugin tries to start a background refresh, it uses {{TS_THREAD_POOL_TASK}}, which crashes when starting the outbound connection because it needs to be a net thread.

> stale_while_revalidate crashing on task threads.
> ------------------------------------------------
>
>                 Key: TS-4790
>                 URL: https://issues.apache.org/jira/browse/TS-4790
>             Project: Traffic Server
>          Issue Type: Bug
>          Components: Core
>            Reporter: James Peach
>            Assignee: James Peach
>             Fix For: 7.0.0
>
>          Time Spent: 0.5h
>  Remaining Estimate: 0h
>
> Seeing this crash:
> {noformat}
> Signal Status:
> siginfo.si_signo:   11 (Segmentation fault)
> siginfo.si_pid:     56                  siginfo.si_uid:     0
> siginfo.si_code:    0x1 (1)             siginfo.si_addr:    0x0000000000000038
> No object mapped at address 0x0000000000000038
> CPU Registers:
> R8 :0x0000000000000000 R9 :0x0000000057c0617a R10:0x146e656fbbaf1913 R11:0x0000000000000246
> R12:0x00002b3c6c015d80 R13:0x0000000001d1e860 R14:0x146e656fbf42a013 R15:0x0000000000000000
> RDI:0x0000000000000020 RSI:0x0000000000000001 RBP:0x00002b3c691d4010 RBX:0x0000000000000000
> RDX:0x00002b3c694d6a30 RAX:0x00002b3c691d4010 RCX:0x00002b3c5b194320 RSP:0x00002b3c694d5dd0
> RIP:0x000000000074676f EFL:0x0000000000010246 CSGSFS:0x0000000000000033 ERR:0x0000000000000004
> TRAPNO:0x000000000000000e OLDMASK:0x0000000000000000 CR2:0x0000000000000038
> Thread 29117, [ET_TASK 1]:
> 0    0x00000000004acb62 crash_logger_invoke(int, siginfo_t*, void*) + 0x72
> 1    0x00002b3c5af8b330 __restore_rt + (nil)
> 2    0x000000000074676f UnixNetVConnection::startEvent(int, Event*) + 0x3f
> 3    0x000000000076d3a0 EThread::process_event(Event*, int) + 0x120
> 4    0x000000000076e05b EThread::execute() + 0x7fb
> 5    0x000000000076ce2a spawn_thread_internal(void*) + 0x4a
> 6    0x00002b3c5af83184 start_thread + 0xc4
> 7    0x00002b3c5bcb737d clone + 0x6d
> 8    0x0000000000000000 0x0 + 0x6d
> {noformat}
> So it is a NULL-dereference in {{UnixNetVConnection::startEvent}}. I suspect the {{stale_while_revalidate_plugin}}.



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