[Couchbase Capella]: Curl timeout when using couchbases schema

Couchbase Capella
1

When running a curl command to a local couchbase server for rebalance endpoint, it works (given username and password are both couchbase ).

curl  -u couchbase:couchbase -X GET http://127.0.0.1:8091/pools/default/rebalanceProgress

When I uses couchbase capella url (svc-***.cloud.couchbase.com), It does not work. What am I missing ?

2

the svc name, when prefixed with _couchbase._tcp., (or _couchbases._tcp. for tls) can be used to to lookup the SRV DNS records for the nodes in the cluster. Then you can curl to those services that are http/s. Be sure that you have added your client to Allowed IPs on Capella. If you would just like to check connectivity, SDK Doctor is a better option. It does the lookup for you. The SDKs have the default Capella root certificate built-in, I believe that SDK Doctor has it as well.

3

Thanks for your reply.
In fact I am looking to check if the node is in rebalance or not. If it is the case, then I want to set the spring boot couchbase health indicator to down until the rebalance finish.
Can you please give me a full curl command ?

4

@amir_choubani, Here is a link to a Google Doc with steps to get from SRV DNS Record to Node Names.

1 Like
5

Thanks. I will try it

6
  1. A properly defined and sized couchbase cluster should work fine during rebalance.
  2. I don’t believe the Health Check status can be set.
7

/pools/default/rebalanceProgress endpoint returns if the couchbase server is in rebalance or not, Am I wrong ?

In that case, spring boot application throw a timeout exception somewhere saying it is unable to connect to couchbase. Looking at couchbase logs, the reason is the rebalance.

8

Please show the exception and logging.
While the Couchbase SDK may log a timeout during a rebalance, it should be automatically retried on an active node.

9

Here is an example of exception log:

c.c.c.c.e.UnambiguousTimeoutException: GetRequest, Reason: TIMEOUT {"cancelled":true,"completed":true,"coreId":"0x2c73426900000001","idempotent":true,"lastDispatchedTo":"svc-d-node-003.eyzzxl7n9mahkrjl.cloud.couchbase.com","reason":"TIMEOUT","requestId":5393539,"requestType":"GetRequest","retried":14,"retryReasons":["ENDPOINT_NOT_AVAILABLE"],"service":{"bucket":"my_bucket_name","collection":"my_collection_name","documentId":"my_document_id","opaque":"0x524d4e","scope":"_default","type":"kv","vbucket":984},"timeoutMs":2500,"timings":{"totalMicros":2501333}}
	at c.c.c.core.msg.BaseRequest.cancel(BaseRequest.java:184)
	at c.c.client.core.msg.Request.cancel(Request.java:71)
	at c.couchbase.client.core.Timer.lambda$register$2(Timer.java:172)
	at c.c.c.c.d.i.n.u.HashedWheelTimer$HashedWheelTimeout.run(HashedWheelTimer.java:715)
	at c.c.c.c.d.i.n.u.c.ImmediateExecutor.execute(ImmediateExecutor.java:34)
	at c.c.c.c.d.i.n.u.HashedWheelTimer$HashedWheelTimeout.expire(HashedWheelTimer.java:703)
	at c.c.c.c.d.i.n.u.HashedWheelTimer$HashedWheelBucket.expireTimeouts(HashedWheelTimer.java:790)
10

That shouldn’t happen. Can you open case with customer support?

11

I do not know how to open case.
Thank you anyway

12
13

This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.