[Bug 271238] mpr (LSI SAS3816) driver not finding all devices in HP D6020 enclosures

From: <bugzilla-noreply_at_freebsd.org>
Date: Wed, 27 Nov 2024 07:26:04 UTC
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=271238

--- Comment #12 from Peter Eriksson <pen@lysator.liu.se> ---
Correct, it's a workaround. Ideally it would have been constructed as some kind
of blacklist so one could have different override values for different
enclosures but that requires more work and this quick fix is good enough.

Getting HPE to fix the D6020 enclosures is pretty much impossible I'd say given
that it went EOL in 2020 (sure, it theory it's still possible to get service
for them since EOSL isn't until 2026). Latest firmware for them was released in
2020 (v2.74) - which we run.

As long as one limit things to 1-2 enclosures per HBA this problem seems to be
masked and things just "work", it's just when you try to use more that it fails
(if the enclosure claims too few slots). This makes me think that there should
be a more clear warning/error displayed by the driver when it detects something
like this - since it will be really unobvious for a sysadmin when it happens
.-)


This (potential) problem isn't just limited to the D6020 enclosures. I did some
quick testing with a few other SAS enclosures f(rom SuperMicro that we use),
and they seem to give incorrect values too - however in those cases they claim
more slots (43 instead of 24) than they actually have so doesn't trigger this
problem :-)

> # mprutil -u1 show enclosures
> Slots      Logical ID     SEPHandle  EncHandle    Type
>   17    300062b21811f800    0000       0001     Direct Attached SES-2
>   43    50030480216b2c7f    0048       0002     External SES-2
>   43    50030480216b3e7f    0062       0003     External SES-2

-- 
You are receiving this mail because:
You are the assignee for the bug.