[
https://issues.apache.org/jira/browse/COUCHDB-11?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12647517#action_12647517
]
Chris Anderson commented on COUCHDB-11:
---------------------------------------
The behavior seems to have gotten worse lately. Checking on this issue, for me,
when I have CouchDB running on port 5984, and then I try to launch another copy
also on the same port, I get no error at all.
It may be helpful to note that if I start Couch A, and make a request, CouchA
responds. Then when I start Couch B, and request again, Couch A still responds.
OTOH, starting fresh from no couches, if I start Couch A, and then Couch B, and
then make an http request, it goes to Couch B. Oh nevermind it's not that
deterministic. I'll have to do more testing to see what happens (not that it
matters).
The easiest thing would be to make Couch fail fast and clear if it's port is
already taken.
> CouchDB should handle errors when opening port gracefully
> ---------------------------------------------------------
>
> Key: COUCHDB-11
> URL: https://issues.apache.org/jira/browse/COUCHDB-11
> Project: CouchDB
> Issue Type: Bug
> Reporter: Noah Slater
> Priority: Minor
>
> ATM CouchDB crashes loud and hard when it tries to open a port that is not
> available. There should be some error handling code in place that produces
> a nice message like so:
> "Could not open port %d: %s" % (port_number, message)
> Obviously, the above is written in Python - but you get the idea.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
