Redis sessions, cache & queues

luceos

forever check storage/logs for the error or enable debug.

forever

Already processed

luceos

forever seems your redis server isn't running or accessible.

forever

Already processed

luceos

forever 0.0.0.0 should allow access from any host that has access to the redis host. If connections still fail then either the host that runs Flarum is not on the same host as redis, or the host that runs Flarum has no access to the host that runs redis.

forever

Already processed

datitisev

forever Is the error in storage/logs the same as the one you sent a screenshot of earlier (the inability to connect to redis)?

TOWUK_ru

i install it with redis-server, have redis all schedules, but extension manager have infinity loading (2+ hours pending)

Redis server v=7.0.15 sha=00000000:0 malloc=jemalloc-5.3.0 bits=64 build=d81b8ff71cfb150e
redis conf:
bind * -::*
protected-mode no

rui5man

Does this plugin conflict with Scout Search? When I set up Redis sessions, the meisearch configured for Scout Search will not work properly. After uninstalling the plugin, it can work normally again. Is it a problem that I set up?

luceos

rui5man not that I know of.

clarkwinkelmann

rui5man the problem could be related to the use of an async queue and not specifically Redis. Are you sure the queue worker is running? If this extension has an option to only use Redis for cache and not for queue, could you check if there is a problem in that configuration?

rui5man

clarkwinkelmann Thank you for your help. It is true that the problem is with asynchronous queues. I will only use Redis for caching and not for queues. The meisearch configured for Scout Search will now work properly. Am I not configuring the queue correctly, or can it only run in this way?

clarkwinkelmann

rui5man Scout should work with asynchronous queues as well. Make sure you follow the "Queue" instructions in the README file of the Redis extension (I see it's also in the first post of this discussion) to configure the queue worker

maisen20

Hello, can anyone help me and tell me what I am doing wrong? I have a redis cluster which also works, as I also use it in other services. But i get this error message. I have 6 nodes from REDIS_CLUSTER_HOST_1 to REDIS_CLUSTER_HOST_6.

flarum.ERROR: Predis\Response\ServerException: MOVED 7293 10.1.0.9:6379

new Blomstra\Redis\Extend\Redis([
	'host' => getenv('REDIS_CLUSTER_HOST_1'),
    'password' => getenv('REDIS_CLUSTER_PASSWORD_1'),
    'port' => getenv('REDIS_CLUSTER_PORT_1'),
    'database' => getenv('REDIS_CLUSTER_DB_1'),
    'cluster' => true,
    'options' => [
        'cluster' => 'redis'
    ]
])

luceos

maisen20 looks like you are connecting to the read-only slave of redis. Not the master.

maisen20

luceos
I have 6 nodes - 3 master nodes and 3 replica nodes.

Don't i just have to specify all 6 nodes for the redis config? I thought a redis cluster does the “routing” automatically? That should also be the advantage, if a master fails its replica can take over for the master.
Am I getting something fundamentally wrong?

luceos

maisen20 I have little experience with this error, it would probably be better if you can find similar reports on the internet for Laravel. The configuration array is adopted from them, so if a solution exists it should be implemented identically on Flarum.

maisen20

I have been searching for hours and could not solve the problem. When I do it like the laravel docs https://laravel.com/docs/8.x/redis#clusters it doesn't work and it doesn't try to connect via the cluster but just uses the default redis configs 127.0.0.1:6379.

I now have the impression that Blomstra\Redis\Extend\Redis does not respect the settings according to laravel. All env vars are set correctly and when i debug i see the correct env values in the logs. But with following config, it always want to connect to [tcp://127.0.0.1:6379 and i get the error message .

flarum.ERROR: Predis\Connection\ConnectionException: Connection refused [tcp://127.0.0.1:6379]

new Blomstra\Redis\Extend\Redis(
 [
        'options' => [
            'cluster' => "redis",
        ],

        'clusters' => [
            'default' => [
                [
                    'host' => getenv('REDIS_CLUSTER_HOST_1'),
                    'password' =>  getenv('REDIS_CLUSTER_PASSWORD_1'),
                    'port' => getenv('REDIS_CLUSTER_PORT_1'),
                    'database' => getenv('REDIS_CLUSTER_DB_1')
                ]
            ],
        ],
    ]
)

luceos

maisen20 instead of declaring the full redis config from laravel, you need to declare only the one you're using, in your example it would boil down to:

                [
                    'host' => getenv('REDIS_CLUSTER_HOST_1'),
                    'password' =>  getenv('REDIS_CLUSTER_PASSWORD_1'),
                    'port' => getenv('REDIS_CLUSTER_PORT_1'),
                    'database' => getenv('REDIS_CLUSTER_DB_1')
                ]

maisen20

You mean like i wrote in my "first" post about this issue above? I already did that and than i get the error:

flarum.ERROR: Predis\Response\ServerException: MOVED 7293 10.1.0.9:6379

new Blomstra\Redis\Extend\Redis([
	'host' => getenv('REDIS_CLUSTER_HOST_1'),
    'password' => getenv('REDIS_CLUSTER_PASSWORD_1'),
    'port' => getenv('REDIS_CLUSTER_PORT_1'),
    'database' => getenv('REDIS_CLUSTER_DB_1'),
    'cluster' => true,
    'options' => [
        'cluster' => 'redis'
    ]
])

As I understand it, a setting is missing somewhere that it should not return the error “Moved”, but that it should be forwarded there?! According to various reports online, this is how it should work:

'options' => [
        'cluster' => 'redis'
    ]

« Previous Page