Hi Andy,

After thinking a bit more thoroughly about this, I believe event this addition would not have a considerable impact on dealing with the problem that we encountered in CASTOR.
Besides being a rather inelegant and error prone solution as you said, the problem is that most of the client disconnects that happened during this particular incident, were disconnecting while being stalled.
The main problem was not the the client was doing an open and then disconnected while doing it but rather he was stalled and never came back for his request. The real problem is signalling the server that the client being stalled has been killed, which I assume is not trivial at all …
Therefore, I propose to put this aside for the moment while I will try to figure out with the Castor team if there are no better solutions to deal with this on our side. Anyway, this happened in some special circumstances and I don’t believe this was a problem ever since.

Thanks,
Elvin

—
Reply to this email directly or view it on GitHub.

Use REPLY-ALL to reply to list

To unsubscribe from the XROOTD-DEV list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=XROOTD-DEV&A=1