Resource icon

Zend Redis Cache 1.6.0

No permission to download
So from this page: http://redis.io/topics/persistence
Looks like having "appendonly no" has better performance than having "appendonly yes".
But much bigger transactions now.
pVKr9X6.png
 
  • Like
Reactions: Xon
Do I add 'maxmemory-policy allkeys-lru' somewhere to /etc/redis.conf?
Yup.

Are you using a socket to communicate to Redis or TCP? If TCP, try refreshing the page a few times and see if the To/From report anything.

If they aren't, can you check that the cache is actually enabled?

You can ignore this if this is a website no one else is touching, and naturally Redis isn't going to be outputting much.
 
Yup.


Are you using a socket to communicate to Redis or TCP? If TCP, try refreshing the page a few times and see if the To/From report anything.

If they aren't, can you check that the cache is actually enabled?

You can ignore this if this is a website no one else is touching, and naturally Redis isn't going to be outputting much.

Right it's a brand new site I'm starting
 
  • Like
Reactions: Xon
In that case looks good. You might want to adjust the number of replica shards for Elastic Search to zero if you aren't running a cluster (this can be done on the fly) (ref).

I ran curl -XPUT '127.0.0.1:9200/mike_xenforo/_settings' -d '{"index" : {"number_of_replicas" : 0}}' But nothing changed
 
I ran curl -XPUT '127.0.0.1:9200/mike_xenforo/_settings' -d '{"index" : {"number_of_replicas" : 0}}' But nothing changed
Try the following to get an understanding of what is happening:
curl -XGET 'http://localhost:9200/_cluster/health?pretty=true'
curl -XGET http://localhost:9200/_cat/shards

There isn't much point running replicas when you have a single server, all it does is consume more diskspace and CPU.

Otherwise, posting in the XFES forum would be the better venue.
 
I don't agree with that determination. How did you arrive at that?
Snapshot (RDB dump file) has a (potentially) high latency hit which can occur unpredictably. Appending log has an overhead due to extra processing for each instruction and a latency hit when it needs to rewrite the append-file to be smaller.
 
  • Like
Reactions: rdn
@Xon, any thougts about this errors. It happend 2-3 times per day.

Code:
Server Error Log
Error Info
CredisException: protocol error, got ' ' as reply type byte - library/Redis/lib/Credis/Client.php:1058
Generated By: Unknown Account, Today at 03:21
Stack Trace
#0 /home/nginx/domains/pijanitvor.com/public/library/SV/RedisCache/RedisDataRegistry.php(175): Credis_Client->__call('exec', Array)
#1 /home/nginx/domains/pijanitvor.com/public/library/XenForo/Dependencies/Abstract.php(147): XenForo_Model_DataRegistry->getMulti(Array)
#2 /home/nginx/domains/pijanitvor.com/public/library/XenForo/FrontController.php(127): XenForo_Dependencies_Abstract->preLoadData()
#3 /home/nginx/domains/pijanitvor.com/public/index.php(13): XenForo_FrontController->run()
#4 {main}
Request State
array(3) {
  ["url"] => string(47) "https://www.pijanitvor.com/index.php?liveupdate"
  ["_GET"] => array(1) {
    ["liveupdate"] => string(0) ""
  }
  ["_POST"] => array(4) {
    ["_xfRequestUri"] => string(49) "/threads/pomoc-za-vase-zdravstvene-probleme.3744/"
    ["_xfNoRedirect"] => string(1) "1"
    ["_xfToken"] => string(8) "********"
    ["_xfResponseType"] => string(4) "json"
  }
}

Code:
Server Error Log
Error Info
CredisException: read error on connection - library/Redis/lib/Credis/Client.php:1058
Generated By: Unknown Account, Today at 03:21
Stack Trace
#0 /home/nginx/domains/pijanitvor.com/public/library/Redis/lib/Credis/Client.php(600): Credis_Client->__call('select', Array)
#1 /home/nginx/domains/pijanitvor.com/public/library/Redis/Cm/Cache/Backend/Redis.php(144): Credis_Client->select(0)
#2 /home/nginx/domains/pijanitvor.com/public/library/Zend/Cache.php(153): Cm_Cache_Backend_Redis->__construct(Array)
#3 /home/nginx/domains/pijanitvor.com/public/library/Zend/Cache.php(94): Zend_Cache::_makeBackend('Redis', Array, false, false)
#4 /home/nginx/domains/pijanitvor.com/public/library/XenForo/Application.php(761): Zend_Cache::factory('Core', 'Redis', Array, Array)
#5 /home/nginx/domains/pijanitvor.com/public/library/XenForo/Application.php(971): XenForo_Application->loadCache(Object(Zend_Config))
#6 /home/nginx/domains/pijanitvor.com/public/library/XenForo/Application.php(1002): XenForo_Application->lazyLoad('cache', NULL)
#7 /home/nginx/domains/pijanitvor.com/public/library/XenForo/Application.php(1647): XenForo_Application::get('cache')
#8 /home/nginx/domains/pijanitvor.com/public/library/XenForo/Model.php(146): XenForo_Application::getCache()
#9 /home/nginx/domains/pijanitvor.com/public/library/SV/RedisCache/RedisDataRegistry.php(152): XenForo_Model->_getCache(true)
#10 /home/nginx/domains/pijanitvor.com/public/library/XenForo/Dependencies/Abstract.php(147): XenForo_Model_DataRegistry->getMulti(Array)
#11 /home/nginx/domains/pijanitvor.com/public/library/XenForo/FrontController.php(127): XenForo_Dependencies_Abstract->preLoadData()
#12 /home/nginx/domains/pijanitvor.com/public/index.php(13): XenForo_FrontController->run()
#13 {main}
Request State
array(3) {
  ["url"] => string(50) "https://www.pijanitvor.com/index.php?ajax/adAction"
  ["_GET"] => array(1) {
    ["ajax/adAction"] => string(0) ""
  }
  ["_POST"] => array(6) {
    ["action"] => string(4) "view"
    ["id"] => string(1) "5"
    ["position"] => string(23) "ad_above_top_breadcrumb"
    ["_xfRequestUri"] => string(69) "/threads/kratki-izvjestaji-sa-ribolova-more-2013-godina.21075/page-10"
    ["_xfNoRedirect"] => string(1) "1"
    ["_xfResponseType"] => string(4) "json"
  }
}
 
@Sunka I've seen that when disk latency was high-ish on a VM under load.

Try disabling AOF so it doesn't touch the disk, and change the RDB interval to be something like a day or so.
 
Last edited:
High load caused those errors and also one where it can't connect at all to the Redis server (experience).
 
So, I change appendonly yes to appendonly no
And comment this three rows:
#save 900 1
#save 300 10
#save 60 10000

Is that OK?
 
Back
Top Bottom