The ALM-12014 Partition Lost is generated on FusionInsight_HD V100R002C50. The hard disk indicator on the RH5288 server (data node) is red, and the slot where the hard disk is located is not the system disk slot. The faulty hard disk is an independent RAID 0.
1. Log in to the FusionInsight Manager. In the Alarm Details area, obtain the HostName from Location and confirm the FusionInsight node type with the customer. It is confirmed that the node is a data node.
2. In the real-time alarm list, check whether the mount record of the hard disk in the /etc/fstab file of the faulty node is deleted after ALM-12014 Partition Lost is reported. If the mount record of the hard disk is not deleted, edit the /etc/fstab file and manually delete it.
3. Check the status of all disks on the faulty node. Run the ps -ef | grep -v "grep" | grep "diskmgt" command to check whether the diskmgt process exists in the system. If it does not, manual installation is required.
4. In the equipment room, remove the faulty disk and replace it with a new one.
5. Restart the operating system. On the RAID controller card configuration page, clear the cache information of the faulty disk and reconfigure RAID 0. After the configuration is complete, restart the server to access the operating system.
After the operating system starts, the diskmgt process automatically partitions, formats, and mounts disks. This process takes about 30 minutes (The time spent varies with disk capacities). Run the mount command to check whether the disk has been mounted to a directory. If yes, log in to FusionInsight Manager and manually clear ALM-12014 Partition Lost of the hard disk. No further action is required.
Check the following:
You are advised to shut down the operating system, then restart it to configure RAID array. Check RAID 0 for the faulty disk, clear the cache information of the RAID array, and reconfigure RAID 0.
After the disk is replaced, check whether the fault is rectified. In addition, manually clear the alarm.