-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
It's opened already,
we need only details and confirmation.
http://bugs.squid-cache.org/show_bug.cgi?id=3775
Amos believe this is the same bug.
Will explore.
15.01.2015 22:07, FredB пишет:
>
>
>> This is not core file size issue, Fred.
>>
>> In
> This is not core file size issue, Fred.
>
> In SunOS core dumps produces by kernel, with uid=0.
>
> Root has unlimited limits set by default.
>
> On this server was running Squid 2 over three years and problems with
> dumps never occurred.
>
> So, the problem not in system settings.
>
> I'
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
This is not core file size issue, Fred.
In SunOS core dumps produces by kernel, with uid=0.
Root has unlimited limits set by default.
On this server was running Squid 2 over three years and problems with
dumps never occurred.
So, the problem not
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
What I observed myself:
This event correlates with users network activity and massive cached
content output during peak hours.
In other hours it does not occurs.
I've checked all network infrastructure, configurations and see no stranges.
I can't
> root @ proxyhost /var/core # ulimit
> unlimited
>
This is not related with ulimit -c , please take a look at ulimit -a
"ulimit -c ulimited" in statupt script only set your core file size to
unlimited
___
squid-users mailing list
squid-users@lists
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
He does not even attempt to write the dump.
Just restarts, write to /var/adm/messages and squid.log and restarts.
15.01.2015 21:29, FredB пишет:
>
>> Squid not required unlimited limits to startup and normal work. In my
>> SMF startup method uses 1
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Solaris 10. Initially squid runs by root. Squid 3 cannot be run via su -
squid -c "".
root @ proxyhost /var/core # ulimit
unlimited
In /var mountpoint I have 13 Gb free space:
rpool/ROOT/s10x_u11wos_24a/var 571M 13.3G 571M /var
:)
I'm neit
> Squid not required unlimited limits to startup and normal work. In my
> SMF startup method uses 128K values to limit filedescriptors during
> startup. System hard limits are above this value.
>
ulimit -c is only about core
___
squid-users mailing l
>
> -BEGIN PGP SIGNED MESSAGE-
> Hash: SHA1
>
> Absolutely sure.
>
> root @ proxyhost /var/core # coreadm
> global core file pattern: /var/core/core.%f.%p
> global core file content: default
>init core file pattern: /var/core/core.%f.%p
>init core file content
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
15.01.2015 21:15, FredB пишет:
>
>
>
>> 2015/01/06 15:52:10 kid1| storeLateRelease: released 0 objects
>> 2015/01/06 15:52:11 kid1| assertion failed: comm.cc:178:
>> "fd_table[conn->fd].halfClosedReader != NULL"
>> 2015/01/06 15:52:15 kid1| Set Curr
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Absolutely sure.
root @ proxyhost /var/core # coreadm
global core file pattern: /var/core/core.%f.%p
global core file content: default
init core file pattern: /var/core/core.%f.%p
init core file content: default
g
> 2015/01/06 15:52:10 kid1| storeLateRelease: released 0 objects
> 2015/01/06 15:52:11 kid1| assertion failed: comm.cc:178:
> "fd_table[conn->fd].halfClosedReader != NULL"
> 2015/01/06 15:52:15 kid1| Set Current Directory to /var/core
>
> Core dump not produced.
>
Are you sure about the right
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Hi gents,
I observe assertion. Time to time, with irregular interval and witout
visible reasons:
2015/01/06 15:52:10 kid1| storeLateRelease: released 0 objects
2015/01/06 15:52:11 kid1| assertion failed: comm.cc:178:
"fd_table[conn->fd].halfClosedR
13 matches
Mail list logo
