Riak-CS False 503 Service Unavailable Error on Client

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Riak-CS False 503 Service Unavailable Error on Client

Valenti, Anthony

We are having a lot of 503 Service Unvailable errors for 1 particular application client(s) when connecting to Riak-CS.  Everything looks fine in Riak/Riak-CS and when I check the Riak-CS access logs, I can see access from other applications to other buckets before, during and after the reported error time.  We have a 5 node cluster that are load balanced and all of them seem to be operating normally and they should be able to handle the incoming connections.  We did find this Jira in Github which looks like our exact problem (https://github.com/basho/riak_cs/issues/1283) , but there is 1 comment and it was closed and I’m not sure of the fix/workaround/resolution.  Has this been resolved in a later version than we are using – riak cs 1.5.3-1?  Is there a way to correct the issue from the client side?

 

Thanks,

Anthony  


_______________________________________________
riak-users mailing list
[hidden email]
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
Reply | Threaded
Open this post in threaded view
|

Re: Riak-CS False 503 Service Unavailable Error on Client

Luke Bakken
Hi Anthony,

Looking at the linked issue it appears that the 503 response can be
returned erroneously when communication between a Riak CS and Riak
node has an error ("If the first member of the preflist is down").

Is there anything predictable about these errors? You say they come
from 1 client only?
--
Luke Bakken
Engineer
[hidden email]


On Tue, Nov 1, 2016 at 7:51 AM, Valenti, Anthony
<[hidden email]> wrote:

> We are having a lot of 503 Service Unvailable errors for 1 particular
> application client(s) when connecting to Riak-CS.  Everything looks fine in
> Riak/Riak-CS and when I check the Riak-CS access logs, I can see access from
> other applications to other buckets before, during and after the reported
> error time.  We have a 5 node cluster that are load balanced and all of them
> seem to be operating normally and they should be able to handle the incoming
> connections.  We did find this Jira in Github which looks like our exact
> problem (https://github.com/basho/riak_cs/issues/1283) , but there is 1
> comment and it was closed and I’m not sure of the fix/workaround/resolution.
> Has this been resolved in a later version than we are using – riak cs
> 1.5.3-1?  Is there a way to correct the issue from the client side?
>
>
>
> Thanks,
>
> Anthony
>
>
> _______________________________________________
> riak-users mailing list
> [hidden email]
> http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com
>

_______________________________________________
riak-users mailing list
[hidden email]
http://lists.basho.com/mailman/listinfo/riak-users_lists.basho.com