I'm maintaining two clusters of Isilon IQ12000x servers running OneFS Version: 7.1.0.2 for a few years and, for the most, its been pretty solid and I'm really happy with them
We have paid the service fees for years and we've had no call outs ever! Only two drives replaced
As they were end of life, but fully serviceable, we did not want to spend the hundreds of thousands to replace them, and now we are unable to pay for support of perfectly usable hardware.
We'll not replace them and the rational is that:
- they are only used as file shares
- we don't use any smart features
- the cluster is mirrored to another large disk store and we have a DR plan in place for recovery should we need it
As for spares we have three entire chassis to raid hardware for and its keep us going for years now.
What we are up against now is that we've got two entire nodes that we cant boot due to corruption and could simply be recovered with the USB image.
One wont boot, the other I tried to reformat but that failed due to having a bad image
This image is not available anywhere, for whatever reason
I know public sharing on a forum is most likely not allowed but i wondered if anyone would be able to point me to it?
It'd be a crying shame to skip these units solely due to this restriction and any help would be appreciated.
Test Journal exited with error - Unable to get battery ship mode.
Checking Isilon Journal integrity...
The node does not have the expected NVRAM device /dev/mnv0. Contact Isilon Customer Support immediately.
Please contact Isilon Customer Support at 1-877-2-ISILON.
Command Options:
1) Enter recovery shell
2) Continue booting
3) Reboot
option>
Enter full pathname of shell or RETURN for /bin/sh:
# isi_reformat_node
Reformatting node and destroying filesystem
Are you sure (yes/no) [no]? yes
This could destroy data
Are you really sure (yes/no) [no]? yes
No handlers could be found for logger "imdd.devices"
Creating lkg from current root
Creating var lkg from current var
/ifs does not appear to be mounted.
Detached
Unable to format
/dev/imdd5a: 512.0MB (1048576 sectors) block size 8192, fragment size 1024
using 12 cylinder groups of 45.13MB, 5777 blks, 11584 inodes.
super-block backups (for fsck -b #) at:
144, 92576, 185008, 277440, 369872, 462304, 554736, 647168, 739600, 832032,
924464, 1016896
Proceeding with v16 newfs...
open: No such file or directory
newfs_efs: newjournal failed
newfs_efs has failed in isi_reformat_node
I've also attached the boot logs if anyone might have another suggestion could suggest another ploy