Launching soon: The world's first vendor agnostic bug scrubLearn more & join waitlist
OPERATIONAL DEFECT DATABASE
...
...
Document Version Release Date Details 2 06/15/2018 Revised the title and updated the Resolution section. 1 04/10/2018 Original Document An issue has been identified with HPE Hitachi Dynamic Link Manager (HDLM) for Linux version 8.1.4-00 to earlier than 8.6.1-00 when attempting to migrate from a boot disk environment to a single-path configuration where the OS may not restart during the process. This issue occurs if HDLM, which gets removed as part of the migration, incorrectly retains an HDLM device registered in the newly created initial RAM disk image file used when starting from a SCSI device. As a result, restart of the OS may not complete after displaying the output prompt, dracut:/#. When creating an initial RAM disk image file (for SCSI), as part of the udev function, there must be a symbolic link from the directory, /dev/disk/by-uuid, to an HDLM device. If the path is read without the presence of a symbolic link to an HDLM device, an incorrect initial RAM disk image file will be created. When the OS restarts using this incorrect initial RAM disk image file, the system searches the file for the UUID corresponding to the HDLM device, but since an incorrect HDLM device is specified, no corresponding UUID is found, which causes wait processing to start. A timeout then occurs because the HDLM device does not exist, and the OS will not start.
Products affected by this issue are as follows: HDLM for Linux 8.1.4-00 to earlier than 8.6.1-00 SUSE LINUX Enterprise Server 12
This issue is resolved in HDLM for Linux version 8.6.1-00 released on 4 June 2018. NOTE: If migration has not been attempted yet, skip to step 9 for an alternate method of creating the initial RAM disk image file. To recover from this issue, follow these steps using the rescue system to re-create an initial RAM disk image file: Start the host by using the rescue system. Start the rescue system from the installation media of the SUSE Linux Enterprise Server. Start the shell by using the rescue system, and then perform the following steps, 3 to 10. Identify the root partitions to be recovered. To do this, use the lsblk command or another command (referred to in step 1). For details on the command you use to identify the root partitions on the rescue system, see the documentation for the OS you are using. Mount the root partitions that you identified in step 3, # mount /dev/{VG name}/{LV name} /mnt Mount the /proc, /dev, and /sys partitions, # mount -t proc none /mnt/proc # mount --rbind /dev /mnt/dev # mount --rbind /sys /mnt/sys Change to the root directory to be recovered, # chroot /mnt /bin/bash Mount the remaining partitions on the system to be recovered, # mount -a In the file /etc/opt/DynamicLinkManager/hdlm.conf, set n for hdlm_dracut . If hdlm_dracut=y is set, use a text editor to change this to hdlm_dracut=n. # cat /etc/opt/DynamicLinkManager/hdlm.conf hdlm_dracut=n Create an initial RAM disk image file, # /usr/bin/dracut --force /boot/initrd-{kernel-version} {kernel-version} Note: If the dracut command is executed on the rescue system, the following message might be output. Failed to look up reference node - returned empty list /lib/kdump/setup-kdump.functions: line 471 /sys/class/net/addr_assing_type: No such file or directory Cannot read permanent address: No such device If the message ( dracut: *** Creating initramfs image file '/boot/initrd-{kernel-version} ' done *** ), which indicates that the initial RAM disk image file has been created, is displayed, there is no issue. Continue with the following steps. Unmount the partitions and then return to the root directory on the rescue system. Then restart the host, # umount -a # exit # reboot After the host starts, re-create the initial RAM disk image file, # /usr/bin/dracut --force /boot/initrd-{kernel-version} {kernel-version} Restart the host. # reboot After the host starts, continue operations from the step, After migration, to the SCSI boot disk environment in the section, "Migration to a single-path boot disk environment" in the Hitachi Command Suite Dynamic Link Manager (for Linux®) User Guide . Continue operations from the step, After the host starts, and check that migration to the SCSI boot disk environment was successful in the section, Migration to a single-path boot disk environment (in the section "Removing HDLM") in the Hitachi Command Suite Dynamic Link Manager (for Linux®) User Guide . When version 8.5.3 is used, perform the steps from step 19 in the section, Migration to a single-path boot disk environment (in the section Removing HDLM) in the Hitachi Command Suite Dynamic Link Manager (for Linux®) User Guide for version 8.5.3 . NOTE: Products sold prior to the November 1, 2015 separation of Hewlett-Packard Company into Hewlett Packard Enterprise Company and HP Inc. may have older product names and model numbers that differ from current models. RECEIVE PROACTIVE UPDATES : Receive support alerts (such as Customer Advisories), as well as updates on drivers, software, firmware, and customer replaceable components, proactively in your e-mail through HPE Subscriber's Choice. Sign up for Subscriber's Choice at the following URL: Proactive Updates Subscription Form.