Resource icon

Zend Redis Cache 1.6.0

No permission to download
Hi Xon,

it would be great, to see on ACP Widget for Redis also the Redis memory usage peak. Thats the only thing I miss.

Ok, and a documentation I understand for a master/slave config on same maschine... ;)
 
it would be great, to see on ACP Widget for Redis also the Redis memory usage peak. Thats the only thing I miss.
This is actually fairly easy, it presents the full contents of the INFO to the xenforo admin template. Add {xen:helper dump, $redis} into the admin template SV_Redis_info to see the full information.

If I have some time, I might make it add all that information and then fiddle with some CSS to hide stuff until you click/hover over it.

Ok, and a documentation I understand for a master/slave config on same maschine... ;)
This is a matter of server configuration
 
I have a Wordpress blog using Redis, how can I make my XF forum use a separate Redis cache?
There are 16 databases by default, Wordpress is using 0.
 
I have a Wordpress blog using Redis, how can I make my XF forum use a separate Redis cache?
There are 16 databases by default, Wordpress is using 0.
IIRC @Xon's addon can't specify the redis db number but wordpress can, so switch it around, Xenforo Redis use db #0 and WP use Redis db #1
 
I have a Wordpress blog using Redis, how can I make my XF forum use a separate Redis cache?
There are 16 databases by default, Wordpress is using 0.
IIRC @Xon's addon can't specify the redis db number but wordpress can, so switch it around, Xenforo Redis use db #0 and WP use Redis db #1
It can. example with all options, not well documented tho.

Try:
Code:
$config['cache']['backendOptions']['database'] = 1;
 
Xon updated Zend Redis Cache with a new update entry:

Maintenance update

  • Option "slave-select" can now be a closure or a string. Default is "preferLocalSlave".
    • String options:
      • preferLocalSlave - fetches IP's via a shell exec
      • preferLocalSlaveLocalDisk - fetches IP's via a shell exec, but caches to /tmp/local_ips
      • preferLocalSlaveAPCu - fetches IP's via a shell exec, but caches to APCu
        • NOT RECOMMENDED
  • New option 'sentinel_persistent' (true/false, uses 'persistent' option if...

Read the rest of this update entry...
 
It would be really useful to flushall from admincp.
This is a bit tricky as some of my add-ons can use different Redis instances, and it is a very heavyweight operation to-do. I'm planning on revising how this add-on presents redis instances to other add-ons before XF2, so I'll look into it :)
 
well you can flush a specific redis database, just make sure addons are using a specific database number ? :)

and/or add config for private vs public redis database /server instances :)
 
@Xon we're also receiving the "This website has encountered an error" issue.

Like the previous member in this thread, we're using redis and the image proxy so that's probably the problem.

Just thought you should know.
 
Without knowing the actual error, this is very impossible to diagnose.

If you are getting lots of connection timeouts or weird protocol errors, please check the VM for High CPU steal.

Run:
Code:
screen
iostat -c -t 30 > /tmp/cpu_stats.log

Then check your max CPU steal after a day or so:

Code:
cat /tmp/cpu_stats.txt | grep -v iowait | awk '{print $5;}'  | sort -rn | more
 
Xon updated Zend Redis Cache with a new update entry:

1.4.0 - maintenance update

This update synchronizes with the Credis and Cm_Cache_Backend_Redis library's upstream, which results in several breaking changes in configuration.

This is a breaking change if you use sentinel support

The following configuration keys have been renamed:
  • sentinel_master_set renamed to sentinel_master
  • slave-select renamed to slave_select_callable
  • sentinel_persistent renamed to...

Read the rest of this update entry...
 
FYI; if you use the connection string format with Redis Sentinel:
Code:
'server' => '127.0.0.1:26379'
This will break with recent php updates.

You must use:
Code:
'server' => 'tcp://127.0.0.1:26379'
or
Code:
'server' => '127.0.0.1', 'port' => 26379,

This will mostly cause a whitepage if this error occurs.
 
Last edited:
@Xon you shoud change in FAQ variable /tmp/cpu_stats.log on both command to log or both to txt


Q. CPU Steal is mentioned as causing latency issue, how do I check this.
A. On Linux, run:

Code:
screen
iostat -c -t 30 > /tmp/cpu_stats.log

Then after a day or so; check your max CPU steal:
Code:
cat /tmp/cpu_stats.txt | grep -v iowait | awk '{print $5;}'  | sort -rn | more

And this is my result.
Is it OK or should I be worried?
Code:
2.02
1.84
1.82
1.71
1.67
1.61
1.53
1.52
1.50
1.48
1.43
1.42
1.41
1.35
1.33
1.30
1.30
1.29
1.29
1.28
1.27
1.27
1.26
1.26
1.25
1.24
 
@Sunka anything under <10% is generally ok, obviously lower is better.

Any spike above 15% and it is basically a case of get a new VM, or migrate the VM onto a new host.
 
Last edited:
Top Bottom