tips, thoughts

Those damn mpt3sas_cm0 messages!

Update: If you’re having these issues, then you’ve probably bought this card from AliExpress or eBay and these cards are bad/damaged…

Linux boasts the best hardware driver support among operating systems (let’s just say some Windows drivers I’ve used caused crashes, slowdowns, or were simply unusable).

When it comes to SCSI Host Bus Adapters (HBAs), all modern cards are typically well-supported. So, if you’re building a machine, there’s a very high chance (almost 99%) that you’ll automatically find a suitable kernel module to initialize and activate your hardware, allowing you to use various software tools for disk management.

However, you might occasionally encounter an issue in the kernel system log (or journal) that looks something like this:

[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)
[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)
[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)
[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)
[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)
[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)
[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)
[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)
[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)
[Sun Jun 9 01:20:59 2024] mpt3sas_cm0: log_info(0x30030109): originator(IOP), code(0x03), sub_code(0x0109)

What do these messages mean? Is there a problem?

Unfortunately, querying the linux-scsi mailing list didn’t yield any helpful information. However, by observing various user-level applications, it appears that some tools use one version of the SCSI API, while others use a different version (let’s call them version 1 and version 2).

With version 1, every interaction with the device (opening communication, sending commands, retrieving data, closing communication) generates these messages. For example, using the drivetemp kernel module with the s-tui tool will produce these messages. Conversely, other tools that communicate directly with the HBA won’t generate them.

In summary, these messages are harmless. If there’s a genuine error, you’ll see a clear indication in the system log or journal.

Exit mobile version