Re: AprMemCache::Put error: Could not find specified socket in poll list. (70015)

[Joshua Marantz]

System load issues can cause transient problems communicating with memcached.  The system should recover within minutes, however.  The pattern I expect to see is that there is a burst of a handful of messages all within a few seconds, then nothing for 30 seconds, because at that point the system has gone into recovery mode, where mod_pagespeed will try not send anything to memcached.

Every 30 seconds, the system will start trying memcached again, until memcached becomes responsive.

Note the timestamps on your messages.  The one you pasted is 10/22 13:40:58.

So if you are looking back through your logs over an extended time on a loaded system,  It's possible to find occasional bursts of memcached issues over a few seconds.  Is this roughly what you are seeing?  Or does the system never recover?

Is it possible that your memcached instance is wedged?  Or that something has happened to the network between your Apache & memcached instances?

-Josh

On Wed, Oct 22, 2014 at 7:47 AM, Rainer Maurer <baufir...@gmail.com> wrote:

Hello,

i get lots of errors in my error log, all the error´s are the same [70015]

2014/10/22 13:40:58 [error] 5315#0: [ngx_pagespeed 1.8.31.4-4009] AprMemCache::Put error: Could not find specified socket in poll list. (70015) on key prop_page/http://www.kontovergleich.jetzt/_VznYLdCCk0@Desktop@dom, value-size 186

 i have absolutely no clue what it is or caused, any help or workaround?

thanks

--
You received this message because you are subscribed to the Google Groups "mod-pagespeed-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mod-pagespeed-di...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mod-pagespeed-discuss/0616c196-bd46-47e8-b83d-2dacc94797ad%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Rainer Maurer]

2014/10/22 12:56:32 [error] 2874#0: [ngx_pagespeed 1.8.31.4-4009] AprMemCache::Put error: Could not find specified socket in poll list. (70015) on key prop_page/http://www.example.com/node/90_VznYLdCCk0@Desktop@dom, value-size 1172014/10/22 13:40:58 [error] 5315#0: [ngx_pagespeed 1.8.31.4-4009] AprMemCache::Get error: Connection refused (111) on key prop_page/http://www.example.com/_VznYLdCCk0@Desktop@beacon_cohort

i got this error too but no others

i use nginx and hhvm 

on this serv i mean i have no memcached running..wondering about the error

service memcached

memcached: unrecognized service

[Rainer Maurer]

i´m just wondering me about the memcached issue...i was sure thats i have it not running

so checking:

xxxx@localhost:~# service memcached

memcached: unrecognized service

i run nginx with hiphop

2014/10/22 12:56:32 [error] 2874#0: [ngx_pagespeed 1.8.31.4-4009] AprMemCache::Put error: Could not find specified socket in poll list. (70015) on key prop_page/http://www.example.com/node/90_VznYLdCCk0@Desktop@dom, value-size 1172014/10/22 13:40:58 [error] 5315#0: [ngx_pagespeed 1.8.31.4-4009] AprMemCache::Get error: Connection refused (111) on key prop_page/http://www.example.com/_VznYLdCCk0@Desktop@beacon_cohort

and i run more then 1 site on this serv. all other sites brings no errors just this one running on worpdress

Am Mittwoch, 22. Oktober 2014 13:53:46 UTC+2 schrieb jmarantz:

System load issues can cause transient problems communicating with memcached.  The system should recover within minutes, however.  The pattern I expect to see is that there is a burst of a handful of messages all within a few seconds, then nothing for 30 seconds, because at that point the system has gone into recovery mode, where mod_pagespeed will try not send anything to memcached.

Every 30 seconds, the system will start trying memcached again, until memcached becomes responsive.

Note the timestamps on your messages.  The one you pasted is 10/22 13:40:58.

So if you are looking back through your logs over an extended time on a loaded system,  It's possible to find occasional bursts of memcached issues over a few seconds.  Is this roughly what you are seeing?  Or does the system never recover?

Is it possible that your memcached instance is wedged?  Or that something has happened to the network between your Apache & memcached instances?

-Josh

On Wed, Oct 22, 2014 at 7:47 AM, Rainer Maurer <baufir...@gmail.com> wrote:

Hello,

i get lots of errors in my error log, all the error´s are the same [70015]

2014/10/22 13:40:58 [error] 5315#0: [ngx_pagespeed 1.8.31.4-4009] AprMemCache::Put error: Could not find specified socket in poll list. (70015) on key prop_page/http://www.kontovergleich.jetzt/_VznYLdCCk0@Desktop@dom, value-size 186

 i have absolutely no clue what it is or caused, any help or workaround?

thanks

--
You received this message because you are subscribed to the Google Groups "mod-pagespeed-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to mod-pagespeed-discuss+unsub...@googlegroups.com.

[Rainer Maurer]

test answer

Am Mittwoch, 22. Oktober 2014 13:53:46 UTC+2 schrieb jmarantz:

System load issues can cause transient problems communicating with memcached.  The system should recover within minutes, however.  The pattern I expect to see is that there is a burst of a handful of messages all within a few seconds, then nothing for 30 seconds, because at that point the system has gone into recovery mode, where mod_pagespeed will try not send anything to memcached.

Every 30 seconds, the system will start trying memcached again, until memcached becomes responsive.

Note the timestamps on your messages.  The one you pasted is 10/22 13:40:58.

So if you are looking back through your logs over an extended time on a loaded system,  It's possible to find occasional bursts of memcached issues over a few seconds.  Is this roughly what you are seeing?  Or does the system never recover?

Is it possible that your memcached instance is wedged?  Or that something has happened to the network between your Apache & memcached instances?

-Josh

On Wed, Oct 22, 2014 at 7:47 AM, Rainer Maurer <baufir...@gmail.com> wrote:

Hello,

i get lots of errors in my error log, all the error´s are the same [70015]

2014/10/22 13:40:58 [error] 5315#0: [ngx_pagespeed 1.8.31.4-4009] AprMemCache::Put error: Could not find specified socket in poll list. (70015) on key prop_page/http://www.kontovergleich.jetzt/_VznYLdCCk0@Desktop@dom, value-size 186

 i have absolutely no clue what it is or caused, any help or workaround?

thanks

--
You received this message because you are subscribed to the Google Groups "mod-pagespeed-discuss" group.

To unsubscribe from this group and stop receiving emails from it, send an email to mod-pagespeed-discuss+unsub...@googlegroups.com.

[Zwilla]

* push