Galaxy
®
GHDX2-ADA DICOM Storage and Archive Appliance Admin Guide
78
Appendix D: Recovery from a Catastrophic Failure
In the event of a catastrophic failure follow the procedures below.
Should any of the following items fail it will be considered a catastrophic failure: memory,
CPU, PSU, ADA drive, SATADOM or RAID Controller. So in most cases any element of the
GalaxyDICOM that fails except the hard drives will be considered a catastrophic failure.
In this case a replacement GalaxyDICOM Appliance (with or without Hard Drives) will be
brought to the site and the old hard drive may be used to retain the Customer’s data.
Failure of the Computer
Failure Discovery
A computer failure will be discovered by the operator because it will stop performing
operations in consistent fashion (if at all). The attempt to power-up will fail (if at all), the
network connections fail, etc. Verify the failure with your hardware support organization
Service Procedures
An overview of the recovery procedure for a Catastrophic Failure is listed below:
1.
Power off the GalaxyDICOM and contact your hardware support organization
personnel for a replacement GalaxyDICOM.
2.
Based on the failure, your hardware support organization will let you know the
procedures to follow for the replacement that will take place.
3.
Disconnect all power and interface cables from the existing GalaxyDICOM.
4.
Based on your individual case, you may be asked to remove hard disk drives from
the failed computer.
5.
Acclimate the replacement GalaxyDICOM to your environment for at least 4 hours
before using.
6.
Replace computer or failed hardware.
7.
Insert hard drives from failed computer if told to do so by your support group.
8.
Reboot new computer.
9.
Reinstall QStar HSM license on the new computer.
10.
Verify file systems consistency (/, /boot, pacs_online, pacs_nearline, postgressdb)
a.
Perform “recoverGalaxyFromRAIDFailure.sh” with the flag “--stop-all”
b.
Perform file system checks using “xfs_check /dev/sdb, sdb2 and sdb3; if
these commands return with no errors continue
c.
Perform “recoverGalaxyFromRAIDFailure.sh” with the flag “--start-all”
11.
Start QStar HSM and verify its health.
12.
Verify operational status of all subsystems.
