[mod_cluster-issues] [JBoss JIRA] (MODCLUSTER-697) Segfault when DeterministicFailover On

[Radoslav Husar (Jira)]

     [ https://issues.jboss.org/browse/MODCLUSTER-697?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Radoslav Husar updated MODCLUSTER-697:
--------------------------------------
    Priority: Blocker  (was: Critical)


> Segfault when DeterministicFailover On
> --------------------------------------
>
>                 Key: MODCLUSTER-697
>                 URL: https://issues.jboss.org/browse/MODCLUSTER-697
>             Project: mod_cluster
>          Issue Type: Bug
>          Components: Native (httpd modules)
>    Affects Versions: 1.3.8.Final
>         Environment: jbcs-httpd24-mod_cluster-native-1.3.8-1.Final_redhat_1.jbcs.el7.x86_64
>            Reporter: Hisanobu Okuda
>            Assignee: Radoslav Husar
>            Priority: Blocker
>
> When DeterministicFailover On, httpd segfaults:
> {code}
> (gdb) bt
> #0  proxy_worker_cmp (a=0x7ffb18002438, b=0x7ffb18002470) at mod_proxy_cluster.c:154
> #1  0x00007ffba4c87ba3 in msort_with_tmp (p=0x7ffb551b37f0, b=0x7ffb551b3748, n=2) at msort.c:124
> #2  0x00007ffba4c87b02 in msort_with_tmp (n=2, b=0x7ffb551b3748, p=0x7ffb551b37f0) at msort.c:45
> #3  msort_with_tmp (p=0x7ffb551b37f0, b=0x7ffb551b3748, n=5) at msort.c:53
> #4  0x00007ffba4c87ef7 in msort_with_tmp (n=5, b=<optimized out>, p=0x7ffb551b37f0) at msort.c:45
> #5  __GI___qsort_r (b=b at entry=0x7ffb18002438, n=n at entry=5, s=s at entry=56, cmp=cmp at entry=0x7ffb90ae5e00 <proxy_worker_cmp>, arg=arg at entry=0x0) at msort.c:254
> #6  0x00007ffba4c88148 in __GI_qsort (b=b at entry=0x7ffb18002438, n=n at entry=5, s=s at entry=56, cmp=cmp at entry=0x7ffb90ae5e00 <proxy_worker_cmp>) at msort.c:308
> #7  0x00007ffb90ae9ae7 in internal_find_best_byrequests (conf=0x55fe55cacb88, node_table=0x7ffb180108e0, context_table=0x7ffb1800fed8, 
>     vhost_table=0x7ffb1802baa0, failoverdomain=0, domain=0x0, r=0x7ffb18006a10, balancer=0x55fe56065140) at mod_proxy_cluster.c:2331
> #8  find_best_worker (balancer=0x55fe56065140, conf=conf at entry=0x55fe55cacb88, r=r at entry=0x7ffb18006a10, domain=domain at entry=0x0, 
>     failoverdomain=failoverdomain at entry=0, vhost_table=vhost_table at entry=0x7ffb1802baa0, context_table=context_table at entry=0x7ffb1800fed8, 
>     node_table=node_table at entry=0x7ffb180108e0, recurse=recurse at entry=1) at mod_proxy_cluster.c:3484
> #9  0x00007ffb90aed611 in proxy_cluster_pre_request (worker=0x7ffb551b3b38, balancer=0x7ffb551b3b30, r=0x7ffb18006a10, conf=0x55fe55cacb88, 
>     url=0x7ffb551b3b40) at mod_proxy_cluster.c:3851
> #10 0x00007ffb9a409fa6 in proxy_run_pre_request () from /usr/jboss/jbossweb/shopapp_ws_m01/modules/mod_proxy.so
> #11 0x00007ffb9a40ec62 in ap_proxy_pre_request () from /usr/jboss/jbossweb/shopapp_ws_m01/modules/mod_proxy.so
> #12 0x00007ffb9a40a7dc in proxy_handler () from /usr/jboss/jbossweb/shopapp_ws_m01/modules/mod_proxy.so
> ...
> (gdb) disassemble 
> Dump of assembler code for function proxy_worker_cmp:
>    0x00007ffb90ae5e00 <+0>:     mov    (%rdi),%rax
> => 0x00007ffb90ae5e03 <+3>:     mov    0x18(%rax),%rdi
> (gdb) info registers rax
> rax            0x9e10fcde4a34ed64       -7056862584431645340
> (gdb) x 0x9e10fcde4a34ed64
> 0x9e10fcde4a34ed64:     Cannot access memory at address 0x9e10fcde4a34ed64
> (gdb)
> {code}



--
This message was sent by Atlassian Jira
(v7.12.1#712002)