Re: Poor read performance on high-end server
Jens Axboe ax...@kernel.dk writes: Also, I didn't see Chris mention this, but if you have a newer intel box you can use hw accellerated crc32c instead. For some reason my test box always loads crc32c and not crc32c-intel, so I need to do that manually. I have a patch for that, will post it later: autoloading of modules based on x86 cpuinfo. -Andi -- a...@linux.intel.com -- Speaking for myself only. -- To unsubscribe from this list: send the line unsubscribe linux-btrfs in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: Poor read performance on high-end server
On 08/08/2010 03:18 AM, Andi Kleen wrote: Jens Axboe ax...@kernel.dk writes: Also, I didn't see Chris mention this, but if you have a newer intel box you can use hw accellerated crc32c instead. For some reason my test box always loads crc32c and not crc32c-intel, so I need to do that manually. I have a patch for that, will post it later: autoloading of modules based on x86 cpuinfo. Great, it is pretty annoying to have to do it manually. Sometimes you forget. And it's not possible to de-select CRC32C and have the intel variant loaded. -- Jens Axboe -- To unsubscribe from this list: send the line unsubscribe linux-btrfs in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Fwd: btrfsck: checksum verify failed
-- Forwarded message -- From: Evert Vorster evors...@gmail.com Date: Sun, Aug 8, 2010 at 1:18 PM Subject: btrfsck: checksum verify failed To: linux-btrfs@vger.kernel.org Hi there. I have a btrfs on a raw device. (/dev/sda insted of in a partition, like /dev/sda1 ) The device in question is a USB hard drive with a 1TB size. The device worked fine for a week or so, and was mounted and unmounted quite a few times without trouble. The last time the device was unmounted it was a normal unmount. Now, when I try to mount the device, I get the following in dmesg: usb 2-1: new high speed USB device using ehci_hcd and address 4 scsi6 : usb-storage 2-1:1.0 scsi 6:0:0:0: Direct-Access Seagate FreeAgent GoFlex 0148 PQ: 0 ANSI: 4 sd 6:0:0:0: Attached scsi generic sg2 type 0 sd 6:0:0:0: [sdb] 1953525167 512-byte logical blocks: (1.00 TB/931 GiB) sd 6:0:0:0: [sdb] Write Protect is off sd 6:0:0:0: [sdb] Mode Sense: 1c 00 00 00 sd 6:0:0:0: [sdb] Assuming drive cache: write through sd 6:0:0:0: [sdb] Assuming drive cache: write through sdb: unknown partition table sd 6:0:0:0: [sdb] Assuming drive cache: write through sd 6:0:0:0: [sdb] Attached SCSI disk device label TeraByte1 devid 1 transid 5352 /dev/sdb parent transid verify failed on 986800398336 wanted 5352 found 5355 parent transid verify failed on 986800398336 wanted 5352 found 5355 parent transid verify failed on 986800398336 wanted 5352 found 5355 btrfs: open_ctree failed Also, if I try to fsck the device, this is the output: b...@dora:~# btrfsck /dev/sdb parent transid verify failed on 986800398336 wanted 5352 found 5355 parent transid verify failed on 986800398336 wanted 5352 found 5355 parent transid verify failed on 986800398336 wanted 5352 found 5355 btrfsck: disk-io.c:739: open_ctree_fd: Assertion `!(!tree_root-node)' failed. Aborted I am using v 0.19 of the btrfs-tools, and kernel version 2.6.34.1 While a little sad that I can't get at my data anymore, I do have backups of it. I also want to use the device again, but want to stop anybody else running into the same problem. So, if you have any requests like maybe a dd of some small part of the filesystem to study and see what went wrong, and maybe how to fix it, I would be happy to oblige. However, If I hear nothing for a day or so, I'll wipe the filesystem and use the device again. Kind regards, -Evert Vorster- -- http://magnatune.com - Music shared the way it should be. -- To unsubscribe from this list: send the line unsubscribe linux-btrfs in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
quick question...
Hi there. Has anybody got a btrfs root filesystem that is running on a bare block device? Would lilo be able to boot an initramfs that is living on such a filesystem? What I intend to do is to erase all the partitions off my system, make a btrfs volume on /dev/sda, and then just use subvolumes in stead of partitions. Is this folly? Would having an initramfs be a boon or a bane in this endeavour? Would lilo writing to the mbr of a device that is claimed by btrfs break the file system? Do you need to have a partition table to have an MBR? Thanks for any answers... -Evert- -- http://magnatune.com - Music shared the way it should be. -- To unsubscribe from this list: send the line unsubscribe linux-btrfs in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: quick question...
i'm not an expert, but ill do my best to answer. replies interspersed below. On Sun, Aug 8, 2010 at 8:57 PM, Evert Vorster evors...@gmail.com wrote: Hi there. Has anybody got a btrfs root filesystem that is running on a bare block device? Would lilo be able to boot an initramfs that is living on such a filesystem? i haven't [ever] used lilo, but AFIAK no bootloader, with the exception of maybe syslinux (extlinux), can boot from a btrfs device; you won't have to worry about the initramfs, because the loader won't even be able to find the kernel. you'll need a boot partition with a filesystem supported by the loader. What I intend to do is to erase all the partitions off my system, make a btrfs volume on /dev/sda, and then just use subvolumes in stead of partitions. Is this folly? not at all. this is what many of us would like to do, but we run into the bootloader issues above. Would having an initramfs be a boon or a bane in this endeavour? a non-issue, as the bootloader cannot even get the kernel (unless your booting the kernel/initramfs from a different device...) Would lilo writing to the mbr of a device that is claimed by btrfs break the file system? i don't think so, but i'm not sure. Do you need to have a partition table to have an MBR? i just read about this recently. IIRC, the partition table is within the 512 byte MBR; it's a 64 byte section starting at byte 446. so... no :-) C Anthony -- To unsubscribe from this list: send the line unsubscribe linux-btrfs in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
