Do I need to add more disks? 🙂
dh0253
15 Posts
April 19, 2023, 3:30 pmQuote from dh0253 on April 19, 2023, 3:30 pmMy CEPH Dashboard has an error:
Does this mean we are out of disk space and we need to add more drives to each PetaSAN node? Our Windows server lost the iSCSI drive and the iSCSI initiator keeps saying "reconnecting" We're trying to push the test environment to see if we can cause any issues before we go to production.
Thank you for the help.
My CEPH Dashboard has an error:
Does this mean we are out of disk space and we need to add more drives to each PetaSAN node? Our Windows server lost the iSCSI drive and the iSCSI initiator keeps saying "reconnecting" We're trying to push the test environment to see if we can cause any issues before we go to production.
Thank you for the help.
admin
2,930 Posts
April 19, 2023, 4:27 pmQuote from admin on April 19, 2023, 4:27 pmIn the majority of cases yes.
unless you have a large inbalance between OSD usage, else best add disk
In the majority of cases yes.
unless you have a large inbalance between OSD usage, else best add disk
dh0253
15 Posts
April 19, 2023, 6:52 pmQuote from dh0253 on April 19, 2023, 6:52 pmI am surprised that the iSCSI stopped working or that the Windows server acknowledged the iSCSI is unavailable and will not reconnect. Usually, when you're out of disk space, Windows will show RED on the drive but will not disconnect the iSCSI. Or it could be that PetaSAN dismounted the iSCSI; however, checking the iSCSI disk status shows they're started and operational. We'll add a 16TB drive to each PetaSAN server and see. Thank you for responding.
I am surprised that the iSCSI stopped working or that the Windows server acknowledged the iSCSI is unavailable and will not reconnect. Usually, when you're out of disk space, Windows will show RED on the drive but will not disconnect the iSCSI. Or it could be that PetaSAN dismounted the iSCSI; however, checking the iSCSI disk status shows they're started and operational. We'll add a 16TB drive to each PetaSAN server and see. Thank you for responding.
admin
2,930 Posts
April 19, 2023, 8:11 pmQuote from admin on April 19, 2023, 8:11 pmThe disk started is at the iSCSI layer. If you have lower level Ceph errors, Cluster Status in Error, you have an issue. Not all Ceph pgs will be in error, any iSCSI i/o to the up pgs will work, the problem happens if the iSCSI tries to access data on down pgs, by design Ceph will wait indefinitely for the pg to respond, but iSCSI will timeout.
The disk started is at the iSCSI layer. If you have lower level Ceph errors, Cluster Status in Error, you have an issue. Not all Ceph pgs will be in error, any iSCSI i/o to the up pgs will work, the problem happens if the iSCSI tries to access data on down pgs, by design Ceph will wait indefinitely for the pg to respond, but iSCSI will timeout.
Last edited on April 19, 2023, 8:11 pm by admin · #4
Do I need to add more disks? 🙂
dh0253
15 Posts
Quote from dh0253 on April 19, 2023, 3:30 pmMy CEPH Dashboard has an error:
Does this mean we are out of disk space and we need to add more drives to each PetaSAN node? Our Windows server lost the iSCSI drive and the iSCSI initiator keeps saying "reconnecting" We're trying to push the test environment to see if we can cause any issues before we go to production.
Thank you for the help.
My CEPH Dashboard has an error:
Does this mean we are out of disk space and we need to add more drives to each PetaSAN node? Our Windows server lost the iSCSI drive and the iSCSI initiator keeps saying "reconnecting" We're trying to push the test environment to see if we can cause any issues before we go to production.
Thank you for the help.
admin
2,930 Posts
Quote from admin on April 19, 2023, 4:27 pmIn the majority of cases yes.
unless you have a large inbalance between OSD usage, else best add disk
In the majority of cases yes.
unless you have a large inbalance between OSD usage, else best add disk
dh0253
15 Posts
Quote from dh0253 on April 19, 2023, 6:52 pmI am surprised that the iSCSI stopped working or that the Windows server acknowledged the iSCSI is unavailable and will not reconnect. Usually, when you're out of disk space, Windows will show RED on the drive but will not disconnect the iSCSI. Or it could be that PetaSAN dismounted the iSCSI; however, checking the iSCSI disk status shows they're started and operational. We'll add a 16TB drive to each PetaSAN server and see. Thank you for responding.
I am surprised that the iSCSI stopped working or that the Windows server acknowledged the iSCSI is unavailable and will not reconnect. Usually, when you're out of disk space, Windows will show RED on the drive but will not disconnect the iSCSI. Or it could be that PetaSAN dismounted the iSCSI; however, checking the iSCSI disk status shows they're started and operational. We'll add a 16TB drive to each PetaSAN server and see. Thank you for responding.
admin
2,930 Posts
Quote from admin on April 19, 2023, 8:11 pmThe disk started is at the iSCSI layer. If you have lower level Ceph errors, Cluster Status in Error, you have an issue. Not all Ceph pgs will be in error, any iSCSI i/o to the up pgs will work, the problem happens if the iSCSI tries to access data on down pgs, by design Ceph will wait indefinitely for the pg to respond, but iSCSI will timeout.
The disk started is at the iSCSI layer. If you have lower level Ceph errors, Cluster Status in Error, you have an issue. Not all Ceph pgs will be in error, any iSCSI i/o to the up pgs will work, the problem happens if the iSCSI tries to access data on down pgs, by design Ceph will wait indefinitely for the pg to respond, but iSCSI will timeout.