Refreshing slots cache infinite loop issue #1766
Comments
|
As it turns out, we get the host and port information through the CLUSTER SLOTS command, but ElastiCache is not resetting correctly and disconnecting because host is a domain address. |
|
If you change the host value in lib/cluster/index.ts on line 863 to get the IP value first if there is an IP value, the above infinite loop problem will be solved. for (let i = 0; i < result.length; ++i) {
const items = result[i];
const slotRangeStart = items[0];
const slotRangeEnd = items[1];
const keys = [];
for (let j = 2; j < items.length; j++) {
if (!items[j][0]) {
continue;
}
this.options.natMap[`${items[j][0]}:${items[j][1]}`] = {
host: items[j][3]?.[1] as string ?? items[j][0],
port: items[j][1]
}
const node = this.natMapper({
host: items[j][0],
port: items[j][1],
});
node.readOnly = j !== 2;
nodes.push(node);
keys.push(node.host + ":" + node.port);
} |
|
Here is a sample of the output when calling the CLUSTER SLOTS command in ElastiCache. |
|
Removing host and port from the tls attribute in the cluster constructor options eliminated the reset infinite loop issue. We need to find the cause of this symptom. this._client = new Redis.Cluster([{
host,
port
}], {
scaleReads: 'slave',
maxRedirections: envManager.atcsRedisClusterMaxRedirections,
retryDelayOnMoved: 100,
redisOptions: {
username: user,
password,
tls: isLocalEnv ? undefined : {
rejectUnauthorized: false
}
}
}) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Environment
Symptoms
When I call any command through the ioredis library, it throws a MOVED error, performs a refreshing slot cache, and succeeds in getting the correct cluster slot information. However, in the reset part, it is not reset correctly, causing a loop in the refreshing slot cache operation.
If the loop continues like that and exceeds the maximum number of times specified in the options, it will exit with a 'Too Many Redirections' error.
Redis Cluster Setup
ioredis debug log
Is anyone else experiencing similar symptoms to me?
The text was updated successfully, but these errors were encountered: