/var/log/messages flooding

[canifer]

Hello, I'm newbie here. I wonder if anyone has facing the same problem like me.

Why does my /var/log/messages are too large? it's reaching 20GBytes!
Here is the footage of its contents

[CODE]

[1911583.244385] EDAC MC0: CE row 1, channel 0, label "CPU#0Channel#1_DIMM#0": Corrected error (Socket=0 channel=1 dimm=0)
[1911583.244394] EDAC MC0: CE row 1, channel 0, label "CPU#0Channel#1_DIMM#0": Corrected error (Socket=0 channel=1 dimm=0)
[1911583.244403] EDAC MC0: CE row 1, channel 0, label "CPU#0Channel#1_DIMM#0": Corrected error (Socket=0 channel=1 dimm=0)
[1911583.244411] EDAC MC0: CE row 1, channel 0, label "CPU#0Channel#1_DIMM#0": Corrected error (Socket=0 channel=1 dimm=0)
[1911583.244420] EDAC MC0: CE row 1, channel 0, label "CPU#0Channel#1_DIMM#0": Corrected error (Socket=0 channel=1 dimm=0)
[1911583.244430] EDAC MC0: CE row 1, channel 0, label "CPU#0Channel#1_DIMM#0": Corrected error (Socket=0 channel=1 dimm=0)
[1911583.244438] EDAC MC0: CE row 1, channel 0, label "CPU#0Channel#1_DIMM#0": Corrected error (Socket=0 channel=1 dimm=0)

I'm using Intel(R) Xeon(R) CPU E5504 and Fedora 17 as Operating System FYI

and I'd like to know is it safe if i delete that file?

Sorry for my English and thanks in advance

[Adunaic]

Hello. It is just a log file, to let you know what is going on. It is safe to delete, but it will just fill up again. I think you need to try and find teh source of the problem.

[DBelton]

Those errors tell you that it found an error (but was able to correct it) in your memory.

It is a possibility that you have a bad memory module, or it needs to be reseated, etc...

Try running a full memtest on it and see what yo come up with.

Under normal circumstances, your /var/log/messages file shouldn't get to be very large, but if you are continuously kicking out errors, then it grows pretty fast. (and yes, 20GB is HUGE for that file)

[canifer]

Originally Posted by Adunaic

Hello. It is just a log file, to let you know what is going on. It is safe to delete, but it will just fill up again. I think you need to try and find teh source of the problem.

Originally Posted by DBelton

Those errors tell you that it found an error (but was able to correct it) in your memory.

It is a possibility that you have a bad memory module, or it needs to be reseated, etc...

Try running a full memtest on it and see what yo come up with.

Under normal circumstances, your /var/log/messages file shouldn't get to be very large, but if you are continuously kicking out errors, then it grows pretty fast. (and yes, 20GB is HUGE for that file)

Thanks for your reply, if I had to turn off fedora for run full memory test, I can't because it's an active server. What should I do? is there anything out there to launch full memory test without rebooting?

Thanks

[DBelton]

I certainly would be looking at scheduling a little down time for that server.

It's able to correct the memory errors so far (at least from what I can tell above) but there is no telling how long it will be until it starts getting errors that it can't correct. Then you can end up with corrupted data, or the server could crash depending upon what is corrupted.

If it's a memory module not seated good, or a bad memory module, you will have to bring it down to fix it anyway.

[canifer]

Originally Posted by DBelton

I certainly would be looking at scheduling a little down time for that server.

It's able to correct the memory errors so far (at least from what I can tell above) but there is no telling how long it will be until it starts getting errors that it can't correct. Then you can end up with corrupted data, or the server could crash depending upon what is corrupted.

If it's a memory module not seated good, or a bad memory module, you will have to bring it down to fix it anyway.

Thanks for your advice DBelton, I should tell my manager about that in order to got permission to do that.