- apache error logs, nothing very useful
- moodle error logs (11 instances running on the server) - nothing consistent at or around the time of the server going down
- system logs - nothing consistent or error reporting at or around the time that Apache is resetting.
I did make some changes to the Apache httpd.conf file recently, using an notice reported in the Apache logs, but that has not stopped the behavior.
I am looking for things on a schedule, since the server is reporting to go down at about the same time.
What in the environment is on a schedule?The virus scan software is on a schedule, and Apache reports a poss. conflict with it on a port, but I have already investigated that possibility. The only other thing I can point to that is on a schedule is the automated backing up of the DBs. The DBs are not on the same server, but are serving the moodle instances on the Apache server in question.
Could the scheduled tasks backing up the DBs on a server B impact the HTTP (apache service) on server A?
The interesting this that I discovered this am, is a couple schedule tasks on server B that have never run, but are on a schedule. I was messing around a couple weeks ago with this, tinkering with a different backup job for a couple DBs that had less usage, did not need to back up daily, so I stuck them into a task and scheduled to run only once a week.I discovered them by going Start|Programs|Accessories|System Tools|Scheduled tasks
I deleted both of the tasks whose status was Could not start.
I also editing the AT1.bat file and REMMED out the lines that are doing the backup. If the server stops going up and down over the next couple days, I will know that the problem is somehow related to the DB backups, URRRRRRRRR.
No comments:
Post a Comment