Hi everyone,

Nice to be here and thanks everyone for their support to couchdb.

We have a installation with couchdb 1.2.2 (a few gigabytes and hundreds of
thousands of documents), with 3 machines being replicated.

Our issue is that the replication from one machine to the other two will
never get to 100% completion, and it will stop somewhere 10-30 documents
below 100% and not change from there for quite a long time.

On the log of the source machine most of what I see is something like this:

*Retrying POST request to https://destination
<https://destination>*****/_bulk_docs in XX seconds due to error
req_timedout*

On the destination machine, we don't see anything special, except a very
high CPU consumption.

Is this a known issue or something upgrade can solve? We have tried to
replicate from source instead of from destination without much luck.

Any help will be appreciated

Thank you

Reply via email to