Thomas Adam wrote:
--- Thomas Adam
wrote: > --- Rob Keeling wrote: We keep getting Dead parent cache errors, with all internet access blocked (as we don`t allow direct connections), this lasts for a minute or so and then clears. I have tried increasing the parent cache timeouts with no sucess.
The question is what am I doing wrong?
You did not attach the error message. My powers of telepathy are weak at the best of time, but guessing games I cannot play.
Dead Parent cache can mean anything, since it relates to a number of issues. I suspect though that it might be the result of a protcol and/or a firewall issue, assuming you have one.
Hmm, seems half my message was truncated. I did go on to say....
If you use a "cache_peer", which becomes unreachable, then you'll get that error message, Rob. Normally though, this can happen when the parent becomes overloaded. If this is the case then you'll need a second parent.
Good luck getting one of these if your connectivity is through a RBC.
If the parent you're using does not use ICP, then you *will* need to set the "no-query" option for the cache_peer.
If the parent cache is actually several machines on "round-robin" DNS you probably will want to disable ICP. Unless you explicitally configure each one by IP address as a peer and hope that the RBC dosn't go and change the IP. -- Mark Evans St. Peter's CofE High School Phone: +44 1392 204764 X109 Fax: +44 1392 204763