Hi, I have a similar question when testing. When I restarted my OSD server, the OSD cannot join the OSD pool automatically. I have to login into the OSD server and type "cosd -i #"(# is the number of OSD).
My question is, can monitor or MDS automatically detect OSDes, other MDSs or monitors and add them into the pool when they restart? Thanks! 在 2011年3月17日下午3:20，Upendra Moturi <[email protected]> 寫道： > Hi > > I too had the same problem. > Only one option i found(just for testing)i had to stop the cluster and > start it again. > > /etc/init.d/ceph start osd<n> does not do anything as this not even > going into start case . > > 2011/3/17 Upendra Moturi <[email protected]> >> >> I too had the same problem. >> Only one option i found(just for testing)i had to stop the cluster and start >> it again. >> >> /etc/init.d/ceph start osd<n> does not do anything as this not even going >> into start case . >> >> >> 2011/3/17 Samuel Just <[email protected]> >>> >>> From the looks of the fdisk output, it seems that the disk was not >>> recognized by the OS when it was reinserted? The OSD daemon won't be >>> able to start if the disk is not found. >>> -Sam >>> >>> On 03/16/2011 05:08 PM, huang jun wrote: >>> > Hi， >>> > Recently, I try to do "adding/removing osds " in ceph wiki >>> > Here is what I ave done >>> > <1> build a cluster with mon/mds/osd0 in the same machine "host0" >>> > And start it,everything is ok >>> > <2> follow the wiki ceph ,add a disk (/dev/sdb4)into host0 >>> > It's ok ,"ceph osd dump -o - "show that there are two osds and >>> > state are in/up >>> > <3> write a 4G file >>> > pull out the osd1 when it about writes 2G ,the osd1 was marked down >>> > and 5 mins laster >>> > it was marked out,but use "df -hl " >>> > I find "/dev/sdb4 885G 2.7M 885G 1% /data/osd1" (it's right?) >>> > so when re-insert the disk ," fdisk -l" show me: >>> > >>> > Disk /dev/sda: 1000.2 GB, 1000204886016 bytes >>> > 255 heads, 63 sectors/track, 121601 cylinders >>> > Units = cylinders of 16065 * 512 = 8225280 bytes >>> > Disk identifier: 0x000d6542 >>> > Device Boot Start End Blocks Id System >>> > /dev/sda1 * 1 3263 26210016 83 Linux >>> > /dev/sda2 3264 6134 23061307+ f W95 Ext'd (LBA) >>> > /dev/sda3 6135 121601 927488677+ 83 Linux >>> > /dev/sda5 3264 4307 8385898+ 82 Linux swap / >>> > Solaris >>> > /dev/sda6 4308 5612 10482381 83 Linux >>> > /dev/sda7 5613 6134 4192933+ 83 Linux >>> > >>> > Disk /dev/sdc: 1000.2 GB, 1000204886016 bytes >>> > 255 heads, 63 sectors/track, 121601 cylinders >>> > Units = cylinders of 16065 * 512 = 8225280 bytes >>> > Disk identifier: 0x000549cb >>> > Device Boot Start End Blocks Id System >>> > /dev/sdc1 * 1 3263 26210016 83 Linux >>> > /dev/sdc2 3264 6134 23061307+ f W95 Ext'd (LBA) >>> > /dev/sdc4 6135 121601 927488677+ 83 Linux >>> > /dev/sdc5 3264 4307 8385898+ 82 Linux swap / >>> > Solaris >>> > /dev/sdc6 4308 5612 10482381 83 Linux >>> > /dev/sdc7 5613 6134 4192933+ 83 Linux >>> > >>> > And finally I can not start osd1 >>> > Because "/etc/init.d/ceph start osd1 " do nothing >>> > Where is my problem?can you find it out for me? >>> > thanks >>> > -- >>> > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in >>> > the body of a message to [email protected] >>> > More majordomo info at http://vger.kernel.org/majordomo-info.html >>> >>> -- >>> To unsubscribe from this list: send the line "unsubscribe ceph-devel" in >>> the body of a message to [email protected] >>> More majordomo info at http://vger.kernel.org/majordomo-info.html >> >> >> >> -- >> Thanks and Regards, >> Upendra.M > > > > -- > Thanks and Regards, > Upendra.M > -- > To unsubscribe from this list: send the line "unsubscribe ceph-devel" in > the body of a message to [email protected] > More majordomo info at http://vger.kernel.org/majordomo-info.html > -- Best Regards, Sylar Shen -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
