This article explains when the Routing-Engine is stuck in the boot loop with the error log "FAILURE - READ_DMA status=51<READY, DSC, ERROR>".

This error occurs when the system attempts a Direct Memory Access (DMA) read operation from the disk, but the operation fails.

A failure in READ_DMA could cause the following issues:

• Routing Engine performance degradation: The RE depends on disk access for log storage, system processes, and configuration management.
• Junos instability: If critical files cannot be read, the Routing Engine may crash or reboot.
• System Failure: In cases of severe disk corruption, the RE might fail to boot properly.

If you get the console access to RE, then the below logs will be observed in a loop:

Enter full pathname of shell or 'recovery' for root password recovery or RETURN for /bin/sh: ACNOTE: to go to multi-user operation, exit the single-user shell (with ^D)
#
#
#ad2: FAILURE - READ_DMA status=51<READY, DSC, ERROR> error=40<UNCORRECTABLE> LBA=46119040. <<<<<<<<<<<<<<<<<
Device 0 stats
direct reqs = 0 requed reqs = 0 retry_fail_reqs = 0 reinit_fail o invalid_device 0 inv_bio_err 0 soft_err = 0 abort_err = 0 ucrc err = 0 dma_err = 0 illen_err = 0 nomedia_err = 0
mcr_err 0 id_err = 0 mc_err = 0 uncorr_err = 0 transxn_err = 0
busy timeout = 0 cmd_issue_err = 0 dma_load_err = 0 dma_start_err = 0
WARNING: system watchdog timer still running,
use /sbin/watchdog -off to disable
Enter full pathname of shell or 'recovery' for root password recovery or RETURN for /bin/sh: ACNOTE: to go to multi-user operation, exit the single-user shell (with ^D)
#
#
#ad2: FAILURE - READ_DMA status=51<READY, DSC, ERROR> error=40<UNCORRECTABLE> LBA=46119040. <<<<<<<<<<<<<<<<<
Device 0 stats

The disk may have bad sectors, which are unreadable sections, or file system corruption, potentially caused by hardware failure. If the disk shows signs of imminent failure or persistent errors, then please open a case with JTAC Support for further investigation and assistance.