This was originally reported against sentry: getsentry/sentry#2530

Issue Type

[x] Bug (Typo)

Steps to Replicate

1. Examine rb/clients.py.
2. Search for shold.

Expected Behaviour

1. Should read should.

Semi-automated issue generated by
https://github.com/timgates42/meticulous/blob/master/docs/NOTE.md

To avoid wasting CI processing resources a branch with the fix has been
prepared but a pull request has not yet been created. A pull request fixing
the issue can be prepared from the link below, feel free to create it or
request @timgates42 create the PR. Alternatively if the fix is undesired please
close the issue with a small comment about the reasoning.

https://github.com/timgates42/rb/pull/new/bugfix_typo_should

Thanks.

Personally I would say MGET is a big deal, pipelines works but maybe it is not as fast as MGET, is it possible that we could support it ?

The idea is simple, when we get one MGET call, we classify keys according to their redis nodes and send the MGET command to the corresponding node, which happens in parallel.

Mostly we use MGET a lot, maybe it is special enough to be supported as an API ?

Environment

use redis-py 5.0.0

Steps to Reproduce

setup a cluster and try to use it

Actual Result

redis version 4.6.0 works

Could you please publish sdist on PyPI?
It is preferred to build FreeBSD port (databases/py-rb) from sdist from PyPI.
Thanks!

I've seen this with operations that accept a score_cast_func keyword argument, such as zrange and it's variants.

For example:

>>> with cluster.all() as client:
...   client.zrange('key', 0, -1)
... 
Traceback (most recent call last):
  File "<console>", line 2, in <module>
  File "/usr/local/lib/python2.7/site-packages/redis/client.py", line 1648, in zrange
    return self.execute_command(*pieces, **options)
TypeError: execute_command() got an unexpected keyword argument 'score_cast_func'

It looks like StrictRedis.zrange invokes FanoutClient.execute_command with withscores and score_cast_func keyword arguments, but execute_command doesn't accept any keyword arguments. It looks this would also affect MappingClient.execute_command but I didn't test that directly. RoutingClient is not affected.

Currently our way to invoke poll is pretty hacky and it might cause us problems in two situations:

• If we write to the buffer limit we might have to read first to get more. I don't see how that is in general possible with redis as redis cannot start processing unless it got the entire pipeline, but i need to verify that this is not an issue with writing to different redises
• I think we might not support a hangup while writing. Need to verify this.

Thanks.

i.e. things like fanout aren't fully documented and its a bit tricky to find source classes in the code

I'm upgrading from an on premise sentry 7.6 installation and blocked by this error. I made a patch similar to this: getsentry/sentry#2322, changing 'path' with 'unix_socket_path' on sentry but trough above exception.

I cannot change the redis server because is being in used for another app so, could you please guide me on how can I implement this change? Thanks in advance!

This can be tested by setting CONFIG SET timeout 1 in Redis, then running:

from time import sleep
from rb import Cluster

cluster = Cluster({0:{'host': '127.0.0.1', 'port': 6379}})

with cluster.map() as c:
    c.get('foo')

sleep(2)

with cluster.map() as c:
    c.get('foo')

Yells with redis.exceptions.ConnectionError: Error while trying to write requests to redis..

The problem is the first poll on the socket is a close event before any data has been written to the socket.

We should detect this and attempt to reconnect.