From denis at ovsienko.info  Fri Aug 30 00:34:39 2024
From: denis at ovsienko.info (Denis Ovsienko)
Date: Thu, 29 Aug 2024 23:34:39 +0100
Subject: cfarm211 is very slow
Message-ID: <20240829233439.1cd3e885@atlas-11>

Hello.

I tried to run a libpcap test build on cfarm211.cfarm.net because that's
the only Solaris 11.3 host with Autoconf >=2.69 I know.  Unfortunately,
anything that has to do with disk access is extremely slow: ls takes
5-10 seconds and ./configure does not complete before I run out of
patience.

Could you have a look and fix it if this is something simple?  If the
problem is not easy to fix, that's fine: I already have results from
other Solaris versions.

Thank you.

-- 
    Denis Ovsienko

From dam at opencsw.org  Fri Aug 30 09:13:47 2024
From: dam at opencsw.org (Dagobert Michelsen)
Date: Fri, 30 Aug 2024 09:13:47 +0200
Subject: cfarm211 is very slow
In-Reply-To: <20240829233439.1cd3e885@atlas-11>
References: <20240829233439.1cd3e885@atlas-11>
Message-ID: <26597F59-81DD-43B5-9848-C74D24CA5C6C@opencsw.org>

Hi Denis,

> Am 30.08.2024 um 00:34 schrieb Denis Ovsienko via buildfarm I tried to run a libpcap test build on cfarm211.cfarm.net because that's
> the only Solaris 11.3 host with Autoconf >=2.69 I know.  Unfortunately,
> anything that has to do with disk access is extremely slow: ls takes
> 5-10 seconds and ./configure does not complete before I run out of
> patience.
> 
> Could you have a look and fix it if this is something simple?  If the
> problem is not easy to fix, that's fine: I already have results from
> other Solaris versions.

One disk from the mirrored zpool failed some time ago and I replaced it
with a spare one I had in the shelve. The resilvering took pretty long
and I thought that the extended I/O times originated from the resilvering
traffic. I just rechecked the latency and that specific disk performs
terribly even after the resilvering has finished, I don?t know why.
I?ll see that I can get a replacement disk but I need to check and
order one which may take a couple of days.

Sorry for the inconvenience! :-(


Best regards

  ? Dago

-- 
"You don't become great by trying to be great, you become great by wanting to do something,
and then doing it so hard that you become great in the process." - xkcd #896


From denis at ovsienko.info  Fri Aug 30 14:38:05 2024
From: denis at ovsienko.info (Denis Ovsienko)
Date: Fri, 30 Aug 2024 13:38:05 +0100
Subject: cfarm211 is very slow
In-Reply-To: <26597F59-81DD-43B5-9848-C74D24CA5C6C@opencsw.org>
References: <20240829233439.1cd3e885@atlas-11>
 <26597F59-81DD-43B5-9848-C74D24CA5C6C@opencsw.org>
Message-ID: <20240830133805.12c1d556@atlas-11>

On Fri, 30 Aug 2024 09:13:47 +0200
Dagobert Michelsen <dam at opencsw.org> wrote:

> that specific disk performs
> terribly

Thank you for looking into this problem.  I've left the build to run
slowly anyway, at this rate if everything goes well it should complete
by the end of today.

The last few times I saw a very similar symptom was servers with
laptop HDDs, and those tend to have firmware that handles a pre-fail
hardware condition by retrying many times to avoid failing the request.
In my case the SMART counters made it clear the HDDs were spent.

-- 
    Denis Ovsienko