Date: Wed, 13 Mar 1996 11:22:36 -0500 (EST) From: Brian Tao <taob@io.org> To: Mark Hittinger <bugs@freebsd.netcom.com> Cc: hackers@FreeBSD.ORG Subject: Barracuda + BT946C + heavy swap == deadlock? (was Re: Web server locks up...) Message-ID: <Pine.BSF.3.91.960313110501.1611I-100000@cabal.io.org> In-Reply-To: <199602171548.JAA03588@freebsd.netcom.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[sorry, just getting back into this thread]
On Sat, 17 Feb 1996, Mark Hittinger wrote:
>
> > From: Joe Greco <jgreco@brasil.moneng.mei.com>
> > To: taob@io.org (Brian Tao)
> > > typically 40 to 60 httpd's running. It exports a 4-gigabyte
> > > filesystem containing access logs to client machines so our customers
> > > can produce statistical reports.
>
> Is the 4 gig drive a Seagate barracuda? (yes for me, bt946c)
Yes, at the time I was using a DEC DSP3053LS and a Seagate ST15150N
also on a BT946C (PCI):
[logs dated Feb 16 21:20:35]
bt0: Bt946C/ 0-PCI/EISA/VLB(32bit) bus
bt0: reading board settings, busmastering, int=10
bt0: version 4.25J, fast sync, parity, 32 mbxs, 32 ccbs
bt0: targ 0 sync rate=10.00MB/s(100ns), offset=15
bt0: targ 2 sync rate=10.00MB/s(100ns), offset=15
bt0: Using Strict Round robin scheme
bt0 at 0x330 irq 10 on isa
bt0 waiting for scsi devices to settle
(bt0:0:0): "DEC DSP3053LS X442" type 0 fixed SCSI 2
sd0(bt0:0:0): Direct-Access 511MB (1046532 512 byte sectors)
sd0(bt0:0:0): with 3117 cyls, 4 heads, and an average 83 sectors/track
(bt0:2:0): "SEAGATE ST15150N 0014" type 0 fixed SCSI 2
sd1(bt0:2:0): Direct-Access 4095MB (8388315 512 byte sectors)
sd1(bt0:2:0): with 3712 cyls, 21 heads, and an average 107 sectors/track
I have since switched to an NCR53c810 with one Segate Medallist
ST51080N 1GB and three Quantum Grand Prix XP34301 4GB drives. Memory has
also been increased from 64MB to 128MB and it handles our FTP/mirror
archives now as well. No silent reboots or machine lockups yet in the 8
days since the switch.
> Do you run alias ip's for 'virtual web sites'? (yes for me, a bunch)
Yes, 34 of them on this machine.
> What ethernet card do you run on the box? (3c509 isa for me)
de0 <Digital DC21140 Fast Ethernet> rev 18 int a irq 12 on pci0:9
de0: DC21140 [10-100Mb/s] pass 1.2 Ethernet address 00:00:c0:39:41:c8
> How large is your swap file? (256mb swap file)
Device 1K-blocks Used Avail Capacity Type
/dev/sd0s1b 131072 7384 123624 6% Interleaved
/dev/sd1s1b 131072 7408 123600 6% Interleaved
Total 262016 14792 247224 6%
> The reason I ask these questions is that other boxes running the same rev
> of FreeBSD will not exhibit the problem at all. I am trying to find the
> common thread.
The disk/controller combination looks suspicious to me. I've got that
4GB Seagate in a non-production mail server right now, but also using an
NCR controller there (back to using the tried-and-true).
> I will see the following kinds of processes hang (unkillable)
> in "D+" state via ps. Innd, Cern httpd, and ps.
>
> Ps seems to have it happen a lot. "ps -ax" will hang whereas simply "ps"
> will not. When "ps -ax" hangs, who and top will run ok.
No comment from me... I can't even get a login on the machine to do
this. :(
--
Brian Tao (BT300, taob@io.org)
Systems Administrator, Internex Online Inc.
"Though this be madness, yet there is method in't"
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?Pine.BSF.3.91.960313110501.1611I-100000>