We were experiencing some issues with VC, and restarted the VC server. We found that it did not resume the server log until after midnight. When looking at the issue more closely, I found that the previous day, the tray client and API log showed errors, and at the same time the memory cleanup routine that you normally see every 5 minutes in the server log stopped running; like the server log logging, the memory cleanup routine didn't resume until after midnight, rather than after the restart.
The tray client error was this, and it cleared up immediately after the VC server restart:
8/17/2008 5:33:39 AM Ping failed, ex: System.NullReferenceException: Object reference not set to an instance of an object.
at 9xipJ.6MQkHMHFWWgxw.6MQkHMHFWWgxw() in C:\sourcefiles\code\VCTray\Connection\ConnectionAPI.vb:line
The API error at the time of failure was this:
8/17/2008 5:32:11 AM IPC send failed, ex: Port is Busy: All pipe instances are busy.
During the VC server restart we saw this for the API log:
8/18/2008 3:10:49 PM IPC send failed, ex: Failed to read from an IPC Port: The pipe has been ended.
8/18/2008 3:11:19 PM IPC send failed, ex: Failed to connect to an IPC Port: The system cannot find the file specified.
The second line "...cannot find file..." has continued to repeat, even after midnight the next day, and everything else looked normal. Tonight (it's about 2:40AM), a new API log has not been started at all so far - from old logs, it looks like this is usually written to continuously.
The questions are:
1) Will the ongoing API error (or lack of activity at all at the moment, at least that we can see by the log) have an impact; and if so, what can we do to fix it?
2) If we notice that VC is not logging to the server log after a VC restart, is there something we can do to get the logging to resume, or will we need to wait until after midnight? Is this connected to the ongoing API error?