Great, I'll try the new code when I get home since the end goal is to use LVM on top of the raid devices and then present the LVs (or /dev/bcach*) to tcm/lio. If there is any debug information I can give you, please let me know. - Justin -----Original Message----- From: Kent Overstreet [mailto:koverstreet@xxxxxxxxxx] Sent: Monday, November 21, 2011 7:54 AM To: Justin Rush Cc: kent.overstreet@xxxxxxxxx; linux-bcache@xxxxxxxxxxxxxxx Subject: Re: can't attach caching device to backing device On Mon, Nov 21, 2011 at 01:51:00PM +0000, Justin Rush wrote: > Yes, the file does exist. Hmm. Weird. I'll look at the code and see if I can think of anything. > Should I try it with the individual devices in the array? Worth a shot, though caching partitions is well tested. > I don't suppose bcache can play with LVM? The version I just pushed should now. > > - Justin > > -----Original Message----- > From: Kent Overstreet [kent.overstreet@xxxxxxxxx] > Received: Sunday, 20 Nov 2011, 11:31pm > To: Justin Rush [jarush@xxxxxxxx] > CC: linux-bcache@xxxxxxxxxxxxxxx [linux-bcache@xxxxxxxxxxxxxxx] > Subject: Re: can't attach caching device to backing device When you > tried to attach the cache, did the file you were echoing to exist? > That looks correct but I'm not sure what else would generate that > error. > > For the UUIDs to show up in blkid, blkid itself would have to be > updated to know about the bcache superblock (IIRC it lives in > util-linux). I should probably ping the author and see if that can get > included. > > If you have the udev rules and probe-bcache installed, the symlinks > /ought/ to show up when you reboot but I haven't tested that in ages. > > I just got the code written so that it ought to be able to work on top > of md devices again, which means I should be able to run it at home > and I'll have to fix probe-bcache if it's broken then :) > > On Mon, Nov 14, 2011 at 3:10 PM, Justin Rush <jarush@xxxxxxxx> wrote: > > Using 3.1.0-rc9 > > > > I created a cache device on /dev/sdk1 # ./make-bcache -C -b512k -w4k > > --writeback /dev/sdk1 > > > > and a backing device on /dev/md127p1. > > ./make-bcache -B /dev/md127p1 > > > > # ./probe-bcache /dev/sdk1 > > 5f42bdd1-4342-4ab6-8dda-855989a09604: UUID="" TYPE="bcache" > > > > # ./probe-bcache /dev/md127p1 > > 6bbac9a4-afbd-4d40-9562-cdd290622c5d: UUID="" TYPE="bcache" > > > > I then registered them by: > > # echo /dev/sdk1 >/sys/fs/bcache/register # echo /dev/md127p1 > > >/sys/fs/bcache/register > > > > I have /dev/bcache0 > > > > When I try to attach the cache, I get this: > > # echo "5f42bdd1-4342-4ab6-8dda-855989a09604" > > >/sys/block/md127/md127p1/bcache/attach > > -bash: echo: write error: No such file or directory > > > > I believe this problem is due to the UUIDs for the two devices, but > > this is just a guess. The UUID for /dev/sdk1 and /dev/md127p1 > > doesn't appear to be registering properly as they don't show up in > > blkid. I tried specifying the UUID to use with the -U option when > > creating the devices, but that didn't seem to help. I do have the > > udev rules file from the tools in place > > > > Does anyone know what is going on here and how to fix it? > > > > Justin Rush > > jarush@xxxxxxxx > > > > > > -- > > To unsubscribe from this list: send the line "unsubscribe > > linux-bcache" in the body of a message to majordomo@xxxxxxxxxxxxxxx > > More majordomo info at http://vger.kernel.org/majordomo-info.html > > > -- > To unsubscribe from this list: send the line "unsubscribe > linux-bcache" in the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html ��.n��������+%������w��{.n�����{���{ay� ʇڙ���f���h������_�(�階�ݢj"��������G����?���&��
