Date: Tue, 21 Oct 2008 09:56:26 -0600 From: "Brad Davis" <brd@freebsd.org> To: "Troy Settle" <troy@psknet.com> Cc: freebsd-isp@freebsd.org Subject: Re: slow file access Message-ID: <35ffa5710810210856q34ec73cbhac8d2bd5b58a780@mail.gmail.com> In-Reply-To: <48F7852D.2080809@psknet.com> References: <48F7852D.2080809@psknet.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On Thu, Oct 16, 2008 at 12:17, Troy Settle <troy@psknet.com> wrote: > > All, > > I have a 4-STABLE box that's been in service for 5+ years. Yes, I know it's > old, but it's been rock-solid, and migration to a new server would take days > to accomplish, but I fear that's what I'm faced with.. > > Specs on the server: Dual Xeon 2.8Ghz / 4GB / ~35GB RAID-1 (system) / ~105GB > RAID-5 (mail storage) > > For the last week or so, I've been getting spurts of MySQL connection > failures in my mail logs, but the total number of connections never even > comes close the max connections allowed for the server. > > Occasionally, our website will appear to completely stall for a few seconds, > and most frustrating of all, when I'm poking around in a shell, things will > sometimes hang for several seconds while performing certain tasks (like cd, > ls, tail, etc...). Even top took like 8 seconds to start displaying a > process list. According to raidutil, both raid volumes are optimal. > Nothing shows up in syslog. Here's some info from top(1): > > last pid: 8799; load averages: 0.20, 0.43, 0.38 up 35+23:48:22 > 14:11:46 > 263 processes: 2 running, 261 sleeping > Mem: 1027M Active, 2100M Inact, 393M Wired, 178M Cache, 199M Buf, 75M Free > Swap: 4096M Total, 72K Used, 4096M Free > > I'm at a loss as to what I should look at next... anyone have any > suggestions before I commit myself to a VERY long weekend moving everything > to a new box? I would check to make sure the disks aren't flaking out. Depending on if you can see the individual disks in the RAID you could run smartctl on them to make sure they aren't remapping sectors or something like that. I have seen this manifest from disk problems where the RAID controller still thinks the disks are OK. Regards, Brad Davis
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?35ffa5710810210856q34ec73cbhac8d2bd5b58a780>