diff for duplicates of <20120503152053.GA7313@localhost.localdomain> diff --git a/a/1.txt b/N1/1.txt index fab6789..5e3f0b5 100644 --- a/a/1.txt +++ b/N1/1.txt @@ -4,27 +4,38 @@ On Thu, May 03, 2012 at 08:17:43AM -0700, Josh Durgin wrote: > > On Fri, Apr 27, 2012 at 01:02:08PM +0200, Christian Brunner wrote: > >> Am 24. April 2012 18:26 schrieb Sage Weil <sage@newdream.net>: > >> > On Tue, 24 Apr 2012, Josef Bacik wrote: -> >> >> On Fri, Apr 20, 2012 at 05:09:34PM +0200, Christian Brunner wrote: -> >> >> > After running ceph on XFS for some time, I decided to try btrfs again. -> >> >> > Performance with the current "for-linux-min" branch and big metadata -> >> >> > is much better. The only problem (?) I'm still seeing is a warning +> >> >> On Fri, Apr 20, 2012 at 05:09:34PM +0200, Christian Brunner wro= +te: +> >> >> > After running ceph on XFS for some time, I decided to try btr= +fs again. +> >> >> > Performance with the current "for-linux-min" branch and big m= +etadata +> >> >> > is much better. The only problem (?) I'm still seeing is a wa= +rning > >> >> > that seems to occur from time to time: > >> > > >> > Actually, before you do that... we have a new tool, -> >> > test_filestore_workloadgen, that generates a ceph-osd-like workload on the -> >> > local file system. It's a subset of what a full OSD might do, but if -> >> > we're lucky it will be sufficient to reproduce this issue. Something like +> >> > test_filestore_workloadgen, that generates a ceph-osd-like workl= +oad on the +> >> > local file system. =A0It's a subset of what a full OSD might do,= + but if +> >> > we're lucky it will be sufficient to reproduce this issue. =A0So= +mething like > >> > -> >> > test_filestore_workloadgen --osd-data /foo --osd-journal /bar +> >> > =A0test_filestore_workloadgen --osd-data /foo --osd-journal /bar > >> > > >> > will hopefully do the trick. > >> > -> >> > Christian, maybe you can see if that is able to trigger this warning? -> >> > You'll need to pull it from the current master branch; it wasn't in the +> >> > Christian, maybe you can see if that is able to trigger this war= +ning? +> >> > You'll need to pull it from the current master branch; it wasn't= + in the > >> > last release. > >> -> >> Trying to reproduce with test_filestore_workloadgen didn't work for -> >> me. So here are some instructions on how to reproduce with a minimal +> >> Trying to reproduce with test_filestore_workloadgen didn't work fo= +r +> >> me. So here are some instructions on how to reproduce with a minim= +al > >> ceph setup. > >> > >> You will need a single system with two disks and a bit of memory. @@ -32,13 +43,16 @@ On Thu, May 03, 2012 at 08:17:43AM -0700, Josh Durgin wrote: > >> - Compile and install ceph (detailed instructions: > >> http://ceph.newdream.net/docs/master/ops/install/mkcephfs/) > >> -> >> - For the test setup I've used two tmpfs files as journal devices. To +> >> - For the test setup I've used two tmpfs files as journal devices.= + To > >> create these, do the following: > >> > >> # mkdir -p /ceph/temp > >> # mount -t tmpfs tmpfs /ceph/temp -> >> # dd if=/dev/zero of=/ceph/temp/journal0 count=500 bs=1024k -> >> # dd if=/dev/zero of=/ceph/temp/journal1 count=500 bs=1024k +> >> # dd if=3D/dev/zero of=3D/ceph/temp/journal0 count=3D500 bs=3D1024= +k +> >> # dd if=3D/dev/zero of=3D/ceph/temp/journal1 count=3D500 bs=3D1024= +k > >> > >> - Now you should create and mount btrfs. Here is what I did: > >> @@ -46,10 +60,13 @@ On Thu, May 03, 2012 at 08:17:43AM -0700, Josh Durgin wrote: > >> # mkfs.btrfs -l 64k -n 64k /dev/sdb > >> # mkdir /ceph/osd.000 > >> # mkdir /ceph/osd.001 -> >> # mount -o noatime,space_cache,inode_cache,autodefrag /dev/sda /ceph/osd.000 -> >> # mount -o noatime,space_cache,inode_cache,autodefrag /dev/sdb /ceph/osd.001 +> >> # mount -o noatime,space_cache,inode_cache,autodefrag /dev/sda /ce= +ph/osd.000 +> >> # mount -o noatime,space_cache,inode_cache,autodefrag /dev/sdb /ce= +ph/osd.001 > >> -> >> - Create /etc/ceph/ceph.conf similar to the attached ceph.conf. You +> >> - Create /etc/ceph/ceph.conf similar to the attached ceph.conf. Yo= +u > >> will probably have to change the btrfs devices and the hostname > >> (os39). > >> @@ -60,38 +77,44 @@ On Thu, May 03, 2012 at 08:17:43AM -0700, Josh Durgin wrote: > >> > >> - Start ceph (e.g. "service ceph start") > >> -> >> - Now you should be able to use ceph - "ceph -s" will tell you about +> >> - Now you should be able to use ceph - "ceph -s" will tell you abo= +ut > >> the state of the ceph cluster. > >> -> >> - "rbd create -size 100 testimg" will create an rbd image on the ceph cluster. +> >> - "rbd create -size 100 testimg" will create an rbd image on the c= +eph cluster. > >> -> > +> >=20 > > It's failing here -> > +> >=20 > > http://fpaste.org/e3BG/ -> -> 2012-05-03 10:11:28.818308 7fcb5a0ee700 -- 127.0.0.1:0/1003269 <== -> osd.1 127.0.0.1:6803/2379 3 ==== osd_op_reply(3 rbd_info [call] = -5 -> (Input/output error)) v4 ==== 107+0+0 (3948821281 0 0) 0x7fcb380009a0 +>=20 +> 2012-05-03 10:11:28.818308 7fcb5a0ee700 -- 127.0.0.1:0/1003269 <=3D=3D +> osd.1 127.0.0.1:6803/2379 3 =3D=3D=3D=3D osd_op_reply(3 rbd_info [cal= +l] =3D -5 +> (Input/output error)) v4 =3D=3D=3D=3D 107+0+0 (3948821281 0 0) 0x7fcb= +380009a0 > con 0x1cad3e0 -> +>=20 > This is probably because the osd isn't finding the rbd class. > Do you have 'rbd_cls.so' in /usr/lib64/rados-classes? Wherever > rbd_cls.so is, -> try adding 'osd class dir = /path/to/rados-classes' to the [osd] +> try adding 'osd class dir =3D /path/to/rados-classes' to the [osd] > section > in your ceph.conf, and restarting the osds. -> -> If you set 'debug osd = 10' you should see '_load_class rbd' in the osd +>=20 +> If you set 'debug osd =3D 10' you should see '_load_class rbd' in the= + osd > log > when you try to create an rbd image. -> +>=20 > Autotools should be setting the default location correctly, but if > you're > running the osds in a chroot or something the path would be wrong. -> +>=20 -Yeah all that was in the right place, I rebooted and I magically stopped getting +Yeah all that was in the right place, I rebooted and I magically stoppe= +d getting that error, but now I'm getting this http://fpaste.org/OE92/ @@ -100,6 +123,7 @@ with that ping thing repeating over and over. Thanks, Josef -- -To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in +To unsubscribe from this list: send the line "unsubscribe linux-btrfs" = +in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html diff --git a/a/content_digest b/N1/content_digest index e265f65..2d72e29 100644 --- a/a/content_digest +++ b/N1/content_digest @@ -21,27 +21,38 @@ "> > On Fri, Apr 27, 2012 at 01:02:08PM +0200, Christian Brunner wrote:\n" "> >> Am 24. April 2012 18:26 schrieb Sage Weil <sage@newdream.net>:\n" "> >> > On Tue, 24 Apr 2012, Josef Bacik wrote:\n" - "> >> >> On Fri, Apr 20, 2012 at 05:09:34PM +0200, Christian Brunner wrote:\n" - "> >> >> > After running ceph on XFS for some time, I decided to try btrfs again.\n" - "> >> >> > Performance with the current \"for-linux-min\" branch and big metadata\n" - "> >> >> > is much better. The only problem (?) I'm still seeing is a warning\n" + "> >> >> On Fri, Apr 20, 2012 at 05:09:34PM +0200, Christian Brunner wro=\n" + "te:\n" + "> >> >> > After running ceph on XFS for some time, I decided to try btr=\n" + "fs again.\n" + "> >> >> > Performance with the current \"for-linux-min\" branch and big m=\n" + "etadata\n" + "> >> >> > is much better. The only problem (?) I'm still seeing is a wa=\n" + "rning\n" "> >> >> > that seems to occur from time to time:\n" "> >> >\n" "> >> > Actually, before you do that... we have a new tool,\n" - "> >> > test_filestore_workloadgen, that generates a ceph-osd-like workload on the\n" - "> >> > local file system. \302\240It's a subset of what a full OSD might do, but if\n" - "> >> > we're lucky it will be sufficient to reproduce this issue. \302\240Something like\n" + "> >> > test_filestore_workloadgen, that generates a ceph-osd-like workl=\n" + "oad on the\n" + "> >> > local file system. =A0It's a subset of what a full OSD might do,=\n" + " but if\n" + "> >> > we're lucky it will be sufficient to reproduce this issue. =A0So=\n" + "mething like\n" "> >> >\n" - "> >> > \302\240test_filestore_workloadgen --osd-data /foo --osd-journal /bar\n" + "> >> > =A0test_filestore_workloadgen --osd-data /foo --osd-journal /bar\n" "> >> >\n" "> >> > will hopefully do the trick.\n" "> >> >\n" - "> >> > Christian, maybe you can see if that is able to trigger this warning?\n" - "> >> > You'll need to pull it from the current master branch; it wasn't in the\n" + "> >> > Christian, maybe you can see if that is able to trigger this war=\n" + "ning?\n" + "> >> > You'll need to pull it from the current master branch; it wasn't=\n" + " in the\n" "> >> > last release.\n" "> >>\n" - "> >> Trying to reproduce with test_filestore_workloadgen didn't work for\n" - "> >> me. So here are some instructions on how to reproduce with a minimal\n" + "> >> Trying to reproduce with test_filestore_workloadgen didn't work fo=\n" + "r\n" + "> >> me. So here are some instructions on how to reproduce with a minim=\n" + "al\n" "> >> ceph setup.\n" "> >>\n" "> >> You will need a single system with two disks and a bit of memory.\n" @@ -49,13 +60,16 @@ "> >> - Compile and install ceph (detailed instructions:\n" "> >> http://ceph.newdream.net/docs/master/ops/install/mkcephfs/)\n" "> >>\n" - "> >> - For the test setup I've used two tmpfs files as journal devices. To\n" + "> >> - For the test setup I've used two tmpfs files as journal devices.=\n" + " To\n" "> >> create these, do the following:\n" "> >>\n" "> >> # mkdir -p /ceph/temp\n" "> >> # mount -t tmpfs tmpfs /ceph/temp\n" - "> >> # dd if=/dev/zero of=/ceph/temp/journal0 count=500 bs=1024k\n" - "> >> # dd if=/dev/zero of=/ceph/temp/journal1 count=500 bs=1024k\n" + "> >> # dd if=3D/dev/zero of=3D/ceph/temp/journal0 count=3D500 bs=3D1024=\n" + "k\n" + "> >> # dd if=3D/dev/zero of=3D/ceph/temp/journal1 count=3D500 bs=3D1024=\n" + "k\n" "> >>\n" "> >> - Now you should create and mount btrfs. Here is what I did:\n" "> >>\n" @@ -63,10 +77,13 @@ "> >> # mkfs.btrfs -l 64k -n 64k /dev/sdb\n" "> >> # mkdir /ceph/osd.000\n" "> >> # mkdir /ceph/osd.001\n" - "> >> # mount -o noatime,space_cache,inode_cache,autodefrag /dev/sda /ceph/osd.000\n" - "> >> # mount -o noatime,space_cache,inode_cache,autodefrag /dev/sdb /ceph/osd.001\n" + "> >> # mount -o noatime,space_cache,inode_cache,autodefrag /dev/sda /ce=\n" + "ph/osd.000\n" + "> >> # mount -o noatime,space_cache,inode_cache,autodefrag /dev/sdb /ce=\n" + "ph/osd.001\n" "> >>\n" - "> >> - Create /etc/ceph/ceph.conf similar to the attached ceph.conf. You\n" + "> >> - Create /etc/ceph/ceph.conf similar to the attached ceph.conf. Yo=\n" + "u\n" "> >> will probably have to change the btrfs devices and the hostname\n" "> >> (os39).\n" "> >>\n" @@ -77,38 +94,44 @@ "> >>\n" "> >> - Start ceph (e.g. \"service ceph start\")\n" "> >>\n" - "> >> - Now you should be able to use ceph - \"ceph -s\" will tell you about\n" + "> >> - Now you should be able to use ceph - \"ceph -s\" will tell you abo=\n" + "ut\n" "> >> the state of the ceph cluster.\n" "> >>\n" - "> >> - \"rbd create -size 100 testimg\" will create an rbd image on the ceph cluster.\n" + "> >> - \"rbd create -size 100 testimg\" will create an rbd image on the c=\n" + "eph cluster.\n" "> >>\n" - "> > \n" + "> >=20\n" "> > It's failing here\n" - "> > \n" + "> >=20\n" "> > http://fpaste.org/e3BG/\n" - "> \n" - "> 2012-05-03 10:11:28.818308 7fcb5a0ee700 -- 127.0.0.1:0/1003269 <==\n" - "> osd.1 127.0.0.1:6803/2379 3 ==== osd_op_reply(3 rbd_info [call] = -5\n" - "> (Input/output error)) v4 ==== 107+0+0 (3948821281 0 0) 0x7fcb380009a0\n" + ">=20\n" + "> 2012-05-03 10:11:28.818308 7fcb5a0ee700 -- 127.0.0.1:0/1003269 <=3D=3D\n" + "> osd.1 127.0.0.1:6803/2379 3 =3D=3D=3D=3D osd_op_reply(3 rbd_info [cal=\n" + "l] =3D -5\n" + "> (Input/output error)) v4 =3D=3D=3D=3D 107+0+0 (3948821281 0 0) 0x7fcb=\n" + "380009a0\n" "> con 0x1cad3e0\n" - "> \n" + ">=20\n" "> This is probably because the osd isn't finding the rbd class.\n" "> Do you have 'rbd_cls.so' in /usr/lib64/rados-classes? Wherever\n" "> rbd_cls.so is,\n" - "> try adding 'osd class dir = /path/to/rados-classes' to the [osd]\n" + "> try adding 'osd class dir =3D /path/to/rados-classes' to the [osd]\n" "> section\n" "> in your ceph.conf, and restarting the osds.\n" - "> \n" - "> If you set 'debug osd = 10' you should see '_load_class rbd' in the osd\n" + ">=20\n" + "> If you set 'debug osd =3D 10' you should see '_load_class rbd' in the=\n" + " osd\n" "> log\n" "> when you try to create an rbd image.\n" - "> \n" + ">=20\n" "> Autotools should be setting the default location correctly, but if\n" "> you're\n" "> running the osds in a chroot or something the path would be wrong.\n" - "> \n" + ">=20\n" "\n" - "Yeah all that was in the right place, I rebooted and I magically stopped getting\n" + "Yeah all that was in the right place, I rebooted and I magically stoppe=\n" + "d getting\n" "that error, but now I'm getting this\n" "\n" "http://fpaste.org/OE92/\n" @@ -117,8 +140,9 @@ "\n" "Josef\n" "--\n" - "To unsubscribe from this list: send the line \"unsubscribe linux-btrfs\" in\n" + "To unsubscribe from this list: send the line \"unsubscribe linux-btrfs\" =\n" + "in\n" "the body of a message to majordomo@vger.kernel.org\n" More majordomo info at http://vger.kernel.org/majordomo-info.html -584c7ca66847c696e2a2b826eda795b50d8273a5572af7f23768cdbb1e4974e1 +9d68a44a3062499c7d0ae470c434256cb078a5bd71ec0dff894fb3570e6a2b1d
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.