Last week I got a message the my /var partition was almost full. Since
it was 10G and I am using lvm2 I added 5G more to it. This evening I
got the same message, now I had to investigate. When I did I found that
/var/log/messages is huge.
# ls -lh /var/log/messages
-rw------- 1 root root 12G Aug 16 20:15 /var/log/messages
It seem to be constantly written to and growing.
# ls -l /var/log/messages
-rw------- 1 root root 11986603021 Aug 16 20:20 /var/log/messages
# ls -l /var/log/messages
-rw------- 1 root root 11986687279 Aug 16 20:20 /var/log/messages
# ls -l /var/log/messages
-rw------- 1 root root 11986759997 Aug 16 20:21 /var/log/messages
# ls -l /var/log/messages
-rw------- 1 root root 11986783403 Aug 16 20:21 /var/log/messages
# ls -l /var/log/messages
-rw------- 1 root root 11986830215 Aug 16 20:21 /var/log/messages
When I tail it I get the following.
# tail /var/log/messages
Aug 16 20:22:59 host kernel: usb-storage: -- transfer complete
Aug 16 20:22:59 host kernel: usb-storage: Bulk command transfer result=0
Aug 16 20:22:59 host kernel: usb-storage: Attempting to get CSW...
Aug 16 20:22:59 host kernel: usb-storage: usb_stor_bulk_transfer_buf:
xfer 13 bytes
Aug 16 20:22:59 host kernel: usb-storage: Status code 0; transferred 13/13
Aug 16 20:22:59 host kernel: usb-storage: -- transfer complete
Aug 16 20:22:59 host kernel: usb-storage: Bulk status result = 0
Aug 16 20:22:59 host kernel: usb-storage: Bulk Status S 0x53425355 T
0x5c33 R 0 Stat 0x0
Aug 16 20:22:59 host kernel: usb-storage: scsi cmd done, result=0x0
Aug 16 20:22:59 host kernel: usb-storage: *** thread sleeping.
What's going on here? Why all these usb-storage messages?
I don't know exactly when this started, but I'm guessing it has to do
with a recent upgrade.
I also thought there's suppose to be some sort of log rotation going on
to prevent this.
Any ideas?
Right now I'm going to "cat /dev/null > /var/log/messages" so I can boot
up in the morning.
Now my /var dropped to 16%.
Thanks,
dhk
08-17-2010, 12:58 AM
Adam Carter
/var/log/messages is huge
Aug 16 20:22:59 host kernel: usb-storage: -- transfer complete
Aug 16 20:22:59 host kernel: usb-storage: Bulk command transfer result=0
Aug 16 20:22:59 host kernel: usb-storage: Attempting to get CSW...
Aug 16 20:22:59 host kernel: usb-storage: usb_stor_bulk_transfer_buf:
xfer 13 bytes
Aug 16 20:22:59 host kernel: usb-storage: Status code 0; transferred 13/13
Aug 16 20:22:59 host kernel: usb-storage: -- transfer complete
Aug 16 20:22:59 host kernel: usb-storage: Bulk status result = 0
Aug 16 20:22:59 host kernel: usb-storage: Bulk Status S 0x53425355 T
0x5c33 R 0 Stat 0x0
Aug 16 20:22:59 host kernel: usb-storage: scsi cmd done, result=0x0
Aug 16 20:22:59 host kernel: usb-storage: *** thread sleeping.
What's going on here? *Why all these usb-storage messages?
You've probably turned on CONFIG_USB_STORAGE_DEBUG when you configured the kernel
08-17-2010, 06:45 AM
Florian CROUZAT
/var/log/messages is huge
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 17 août 2010, at 02:58, Adam Carter wrote:
>> Aug 16 20:22:59 host kernel: usb-storage: -- transfer complete
>> Aug 16 20:22:59 host kernel: usb-storage: Bulk command transfer result=0
>> Aug 16 20:22:59 host kernel: usb-storage: Attempting to get CSW...
>> Aug 16 20:22:59 host kernel: usb-storage: usb_stor_bulk_transfer_buf:
>> xfer 13 bytes
>> Aug 16 20:22:59 host kernel: usb-storage: Status code 0; transferred 13/13
>> Aug 16 20:22:59 host kernel: usb-storage: -- transfer complete
>> Aug 16 20:22:59 host kernel: usb-storage: Bulk status result = 0
>> Aug 16 20:22:59 host kernel: usb-storage: Bulk Status S 0x53425355 T
>> 0x5c33 R 0 Stat 0x0
>> Aug 16 20:22:59 host kernel: usb-storage: scsi cmd done, result=0x0
>> Aug 16 20:22:59 host kernel: usb-storage: *** thread sleeping.
>>
>> What's going on here? Why all these usb-storage messages?
>
> You've probably turned on CONFIG_USB_STORAGE_DEBUG when you configured the kernel
Yeah and not only it makes your logs huge, but it also slows down your box using your CPU to flood while doing USB transfer.
On my old box, with CONFIG_USB_STORAGE_DEBUG set, syslog-ng eats 25% of my CPU when doing ntfs3g.
Totally something you want to turn off, unless you have serious issues with USB.
- ----
Florian
/ For security reasons, all text in this mail
is double-rot13 encrypted. /
On 08/17/2010 02:45 AM, Florian CROUZAT wrote:
>
> On 17 août 2010, at 02:58, Adam Carter wrote:
>
>>> Aug 16 20:22:59 host kernel: usb-storage: -- transfer complete
>>> Aug 16 20:22:59 host kernel: usb-storage: Bulk command transfer result=0
>>> Aug 16 20:22:59 host kernel: usb-storage: Attempting to get CSW...
>>> Aug 16 20:22:59 host kernel: usb-storage: usb_stor_bulk_transfer_buf:
>>> xfer 13 bytes
>>> Aug 16 20:22:59 host kernel: usb-storage: Status code 0; transferred 13/13
>>> Aug 16 20:22:59 host kernel: usb-storage: -- transfer complete
>>> Aug 16 20:22:59 host kernel: usb-storage: Bulk status result = 0
>>> Aug 16 20:22:59 host kernel: usb-storage: Bulk Status S 0x53425355 T
>>> 0x5c33 R 0 Stat 0x0
>>> Aug 16 20:22:59 host kernel: usb-storage: scsi cmd done, result=0x0
>>> Aug 16 20:22:59 host kernel: usb-storage: *** thread sleeping.
>>>
>>> What's going on here? Why all these usb-storage messages?
>
>> You've probably turned on CONFIG_USB_STORAGE_DEBUG when you configured the kernel
>
> Yeah and not only it makes your logs huge, but it also slows down your box using your CPU to flood while doing USB transfer.
> On my old box, with CONFIG_USB_STORAGE_DEBUG set, syslog-ng eats 25% of my CPU when doing ntfs3g.
> Totally something you want to turn off, unless you have serious issues with USB.
>
> ----
> Florian
> / For security reasons, all text in this mail
> is double-rot13 encrypted. /
>
Yes, I have it turned on in the kernel.
# grep CONFIG_USB_STORAGE_DEBUG
/etc/kernels/kernel-config-x86_64-2.6.32-gentoo-r7
CONFIG_USB_STORAGE_DEBUG=y
I don't understand why it just started to be a problem. I've had this
kernel for almost four months. Shouldn't the log's be rotated and purged?
I've been using the default kernel config from the install. I'll make a
new kernel with it turned off.
Thanks,
dhk
08-17-2010, 11:17 AM
Alex Schuster
/var/log/messages is huge
dhk writes:
> I don't understand why it just started to be a problem. I've had this
> kernel for almost four months. Shouldn't the log's be rotated and
> purged?
Depends on your system logger. Metalog does this automatically, but for
others you probably have to install logrotate.
Wonko
08-17-2010, 11:45 AM
dhk
/var/log/messages is huge
On 08/17/2010 07:17 AM, Alex Schuster wrote:
> dhk writes:
>
>> I don't understand why it just started to be a problem. I've had this
>> kernel for almost four months. Shouldn't the log's be rotated and
>> purged?
>
> Depends on your system logger. Metalog does this automatically, but for
> others you probably have to install logrotate.
>
> Wonko
>
>
>
Right, I need logrotate. I though it was installed. Must have been on
the old system.
Thanks,
dhk
08-17-2010, 11:40 PM
Peter Humphrey
/var/log/messages is huge
On Tuesday 17 August 2010 11:57:13 dhk wrote:
> I don't understand why it just started to be a problem. I've had
> this kernel for almost four months.
/var/log/messages is huge
