I have a minimal debian 64bit server with nginx php5-fpm mysql and
xcache installed on it. Seems everything is ok, but as soon as I decided
to use xcache-admin skripts accessing them are ok, it asks the login and
password and accepts the one I put in xcache.ini, but then it kills the
fpm child that must serve the request and quiting with following log
entries
in php-fpm.log
WARNING: [pool www] child 3385 exited on signal 11 (SIGSEGV) after
833.669396 seconds from start
in nginx log
[error] 3040#0: *48 recv() failed (104: Connection reset by peer) while
reading response header from upstream, client: , server: localhost,
request: “GET /xcache-admin/ HTTP/1.1”, upstream:
“fastcgi://127.0.0.1:9000”, host:
WARNING: [pool www] child 3385 exited on signal 11 (SIGSEGV) after 833.669396
seconds from start
Since php segfaults there is nothing nginx can really do.
Can only suggest to enable core dumps to pinpoint the problem and/or
send those to the Xcache developer or try another opcode cacher
like APC (in my own experience after comparing nearly all - APC while
not being always the fastest (for example eAccelerator can
outperform) under load usually has had less problems in the past).
There are also quite a plenty of segfault tickets and one which looks
similar to your: http://xcache.lighttpd.net/ticket/272
So maybe try to set xcache.admin.enable_auth = Off and see if it helps.
rr
This forum is not affiliated to the Ruby language, Ruby on Rails framework, nor any Ruby applications discussed here.