ScriptLoad call ForEachShard, slaves will return io timeout

[Bensonzjy]

why ScriptLoad call ForEachShard -> (masters, slaves), slaves readonly, aws redis oss return io timeout, call ForEachMaster better ?

[ndyakov]

@Bensonzjy thank you for reaching out. Could you please describe your problem, provide source code and version information?

[Bensonzjy]

go-redis version: v9.7.0
go.mod require: github.com/redis/go-redis/v9 v9.7.0
my code:
luaIncrSHA, err := clusterClient.ScriptLoad(ctx, luaIncrScript).Result()
return err != nil, err is io timeout, because exec load script on redis slave server

https://github.com/redis/go-redis/blob/v9.7.0/osscluster_commands.go#L35

[ndyakov]

@Bensonzjy have you checked https://redis.uptrace.dev/guide/go-redis.html#dial-tcp-i-o-timeout ? Are you able to connect to the slaves? In general, the error you are getting is not related to the script load command, but it is related to the client not being able to connect to the slave. Please check your network. Feel free to close this issue.

[Bensonzjy]

my redis cluster info:
cluster info: cluster_state:ok cluster_slots_assigned:16384 cluster_slots_ok:16384 cluster_slots_pfail:0 cluster_slots_fail:0 cluster_known_nodes:2 cluster_size:1 cluster_current_epoch:0 cluster_my_epoch:0 cluster_stats_messages_sent:0 cluster_stats_messages_received:0 total_cluster_links_buffer_limit_exceeded:0

one master and other one slave, The ForEachShard method will traverse all nodes, including read-only slave nodes.

[ndyakov]

@Bensonzjy you are correct that the ForEachShard will traverse all nodes. Script load should be executed on all nodes, here is the output of command info script|load:

127.0.0.1:6379> command info script|load
1)  1) "script|load"
    2) (integer) 3
    3) 1) noscript
       2) stale
    4) (integer) 0
    5) (integer) 0
    6) (integer) 0
    7) 1) @slow
       2) @scripting
    8) 1) "request_policy:all_nodes"
       2) "response_policy:all_succeeded"
    9) (empty array)
   10) (empty array)

Take a look at request_policy. Your issue is probably network related. See how your cluster is configured. Maybe increase the timeouts, since you are receiving timeout issue.