Re: Randomly occurring segmentation fault

Tautvydas wrote: > recently all servers started randomly crashing with apache segmentation > fault. By analyzing core dumps I can see that cause should be my code, > but the problem is that system is way to big to easily find a line witch > crashes servers. Currently there are six servers running in parallel, > but interesting thing, that if crashes one, other five crashes too. > There are some data retrieving from 3rd party sources (using various > technologies like SOAP, XML, or JSON), so it could be the reason why all > servers die all together. > I'm using Apache 2.4.7, PHP 5.5.9 and it's all running on Ubuntu 14.04 > PHP modules enabled: opcache, > pdo,curl,json,memcached,mysqli,mysql,pdo_mysql,realine,redis,igbinary > Apache modules enabled: access_compat.load > alias, auth_basic, authn_core, authn_file, authz_core, authz_host, > authz_user, deflate, dir, env, filter, mime, mpm_prefork, negotiation, > php5, rewrite, setenvif, status > > Basically all I have is this crashdump backtrace: > > #0 0x00007f0259e41621 in zend_hash_quick_find () from /usr/lib/apache2 > modules/libphp5.so > #1 0x00007f0259e58320 in zend_std_write_property () from > /usr/lib/apache2/modules/libphp5.so > #2 0x00007f0259e38495 in add_property_string_ex () from > /usr/lib/apache2/modules/libphp5.so > #3 0x00007f0259cd8f0f in ?? () from /usr/lib/apache2/modules/libphp5.so > #4 0x00007f0259ce76fd in add_soap_fault () from > /usr/lib/apache2/modules/libphp5.so <snip> > Maybe someone have an idea what it is or wheere to dig next? It appears to be a SOAP related problem (see line #4 in the backtrace). Unfortunately, I can't help further. -- Christoph M. Becker