https://bz.apache.org/bugzilla/show_bug.cgi?id=58491
Bug ID: 58491
Summary: Core dumps appear when the load increases
Product: Apache httpd-2
Version: 2.4.16
Hardware: Other
OS: Linux
Status: NEW
Severity: major
Priority: P2
Component: mod_proxy_http
Assignee: [email protected]
Reporter: [email protected]
Hi,
We are currently facing an issue with Apache since our 2.4 upgrade. When the
load increase too much (250 hits/s) the http instance generates core files.
Here is the kind of core that we have :
Core was generated by `/tech/apache/runtime/httpd-2.4.16/bin/httpd -D
inst-switch -d /tech/apache -f /'.
Program terminated with signal 11, Segmentation fault.
#0 apr_pool_cleanup_register (p=0x400000010, data=0x3181c78,
plain_cleanup_fn=0x7f15bb4e7cc0 <thread_cond_cleanup>,
child_cleanup_fn=0x427b10 <apr_pool_cleanup_null@plt>)
at memory/unix/apr_pools.c:2213
2213 memory/unix/apr_pools.c: No such file or directory.
in memory/unix/apr_pools.c
(gdb) bt
#0 apr_pool_cleanup_register (p=0x400000010, data=0x3181c78,
plain_cleanup_fn=0x7f15bb4e7cc0 <thread_cond_cleanup>,
child_cleanup_fn=0x427b10 <apr_pool_cleanup_null@plt>)
at memory/unix/apr_pools.c:2213
#1 0x00007f15bb4e7de8 in apr_thread_cond_create (cond=0x3181c20,
pool=0x3181168) at locks/unix/thread_cond.c:55
#2 0x00007f15bb93f5ec in apr_reslist_create (reslist=0x1e46300, min=<value
optimized out>, smax=<value optimized out>, hmax=25, ttl=<value optimized out>,
con=<value optimized out>,
de=0x7f15b745a3e0 <connection_destructor>, params=0x291d9e8,
pool=0x3181168) at misc/apr_reslist.c:299
#3 0x00007f15b745af2d in ap_proxy_initialize_worker (worker=0x291d9e8,
s=0x1eaaa38, p=0x1e46148) at proxy_util.c:1848
#4 0x00007f15b7455cac in proxy_handler (r=0x7f1570002970) at mod_proxy.c:1087
#5 0x000000000044b210 in ap_run_handler ()
#6 0x000000000044f38e in ap_invoke_handler ()
#7 0x000000000046201a in ap_process_async_request ()
#8 0x000000000046217f in ap_process_request ()
#9 0x000000000045e4ce in ap_process_http_connection ()
#10 0x0000000000456020 in ap_run_process_connection ()
#11 0x00007f15ba0c7f2a in process_socket (thd=0x2fce5f0, dummy=<value optimized
out>) at worker.c:619
#12 worker_thread (thd=0x2fce5f0, dummy=<value optimized out>) at worker.c:978
#13 0x00007f15bac739d1 in start_thread () from /lib64/libpthread.so.0
#14 0x00007f15ba9c08fd in clone () from /lib64/libc.so.6
This http instance is a reverse proxy with the following configuration at
server level :
ProxyPreserveHost On
BalancerPersist On
RewriteRule /(.*) balancer://tomcat7_80/$1 [proxy]
With the balancer defined at the server level :
<Proxy balancer://tomcat7_80>
ProxySet lbmethod=bybusyness maxattempts=1
BalancerMember http://127.0.6.49:80 lbset=0 smax=0 ttl=3
connectiontimeout=2 retry=60
BalancerMember http://10.70.6.48:80 lbset=1 smax=0 ttl=3
connectiontimeout=2 retry=60
</Proxy>
We have the same configuration on live environment with a much higher load
without facing any issue with Apache 2.2.
Details about our platform :
- OS : RHEL 6.4
- Apache 2.4.16 compiled with apr-1.5.2 and apr-util-1.5.4
- MPM_worker
--
You are receiving this mail because:
You are the assignee for the bug.
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
