Strange -- my times stays about the same:
$ time ./xlog aa fsync
real0m23.695s
user0m0.020s
sys 0m0.150s
$ time ./xlog aa fsync
real0m24.024s
user0m0.030s
sys 0m0.130s
$ rm aa
$ time ./xlog aa fsync
real0m23.979s
user0m0.020s
sys
Hello,
This might be a none SCSI problem, but I just discovered a nasty fault in
the 2.4.2 kernel.
When creating a loopback file on this way:
# dd if=/dev/zero of=test.img bs=1k count=1000
# losetup /dev/loop0 test.img
# mkfs -t ext2 /dev/loop0 1000 - here it's hanging and this process got a
Alan Cox wrote:
I think I've found a stability problem in the SCSI subsystem of actual
kernel 2.4.2.
The 2.4.2 adaptec driver has plenty of problems. Either use the -ac version
and/or get Justin Gibbs new driver
-
I would second that and Adaptec deserves some praise for this.
Klaus Naumann wrote:
rakesh rakesh wrote:
Hi List,
I wanted to know in which order different Host
adapters in the system are detected by the SCSI.
You can have a look in linux/drivers/scsi/hosts.c .
It's in there. Why do you need to know that anyway ?
While this is true in the
Hi!
"Jeremy" == Jeremy Hansen [EMAIL PROTECTED] writes:
Jeremy We're doing some mysql benchmarking. For some reason it seems that ide
Jeremy drives are currently beating a scsi raid array and it seems to be related
Jeremy to fsync's. Bonnie stats show the scsi array to blow away ide as
Hello,
Michael Widenius wrote on Monday, March 05, 2001:
I wonder from where the fdatasync() is comming; MySQL is not doing
those (if you are not running mysqld with --flush)
The call is either a fsync or an fdatasync that is done by Berkley DB on the
transaction log.
Regards,
Chris
Any plans to add support or is there support in the 2.4 kernel for
hot-plugging a PCI adapter? It may be in the sources already, but I can't
seem to locate it. Any help would be appreciated.
Thanks,
Duane Grigsby
-
To unsubscribe from this list: send the line "unsubscribe linux-scsi" in
the
Duane Grigsby wrote:
Any plans to add support or is there support in the 2.4 kernel for
hot-plugging a PCI adapter? It may be in the sources already, but I can't
seem to locate it. Any help would be appreciated.
For devices, the support is already there. See Documentation/pci.txt.
Look for
Since the intention of fsync and fdatasync seems to be
to write dirty fs buffers to persistent storage (i.e.
the "oxide") then the best time is not necessarily
the objective. Given the IDE times that people have
been reporting, it is very unlikely that any of those
IDE disks were really doing
9 matches
Mail list logo