Please let me know:
How much expertise do you have in server/linux/microsoft?
How comfortable do you are in change settings on your server?
Is a production server?
how many users connected on this server?
how do you have it connected to computers?
how much disk space do you have available?
Thank you, Jader, for offering to help.
As to your first set of questions...
Most all of my expertise is in the server/microsoft area. (but it only took about 5 minutes of using Linux for me to see the error, and much wasted time, of my ways!)
2) I am very comfortable with working in a command line enviroment. Just not very familiar with Linux... yet.
3&4) Semi-Production. I have 6 people that I have routed through this server in order to help put it to the test. Everyone else is still on my mail server.
5)Setup is like this; Internet entering the SME box, in Gateway/Server mode. Out of SME box and into a wireless router. Ther router is set to use DHCP for all connections.From there, various Desktop and Laptop systems are served.
6) With RAID set as I have it (using 5 9.1GB SCSI drives) a bit over 26GB's available.
I've seen most of internet failures to be squid related.
Even if SMB still works!
VERY OFTEN that's because the squid.log file reaching 2GB limit.
If this is your problem you can stop squid, touch squid.log and restart it.
I will try this. But I'm not sure what you mean by "touch squid.log"? Do you mean for me to have a look in it after stopping squid?
Never mind that last line. I just looked up "touch" and now I get it!
So I came back, edited and edited this note.
After looking through my logs I found instances, that always coincided with the troubles I'm having, in the DNSCACHE/CURRENT.log. These are listed as "ServFails". A search, in the forums, for this turned up the following thread...
http://forums.contribs.org/index.php?topic=25403.0I found no real help on this, though. Any thoughts?
Again, thank you Jader.
Where were you searching ?
The obvious place to start is your log files.
See server manager, view log files, select messages log file
Ray, Thank you too for helping out.
I was searching these forums. I do look through my logs, but what I was searching for was how to setup the command... /sbin/e-smith/signal-event reboot ( which I already knew, but thank you)... into a cron job so that I could have my server reboot at a given time.
You can also view all the other log files to look for clues eg squid & http.
Make a note of what time the problem happens and look in your log files prior to the problem occurring.
I've done this and, as I mentioned above, It seems to be related to the "ServFail" errors.
Has this machine been performing without problem or has it always had this problem ?
Just after the install of SME7 rc2 is the first time for me to run into this problem on this server... or any of my servers, for that matter.
Any ideas/help will be appreciated.
Thank you!
Jeffrey