Date: Sun, 17 Oct 1999 17:09:59 -0400 From: shashi@websi.com To: freebsd-questions@freebsd.org, freebsd-isp@freebsd.org Cc: shashi@websi.com Subject: Hard Disk failure, mail for virtual host misbehaving Message-ID: <19991017170959.A4993@Shift-F1.com>
next in thread | raw e-mail | index | archive | help
Hi,
I have a Pentium III, 376MB, 3x9.1GB SCSI disk box that I colocate at an
ISP. Recently I had problems connecting and found that the boc would keep
rebooting after trying to sync devices. The tech guys went into single user
mode and started basic stuff so I could remote telnet and fix things.
1. I found that SCSI disk #1 was not responding. #0 is ok, #2 is ok, #1 is
gone. I had web and user homes in this disk.
2. I removed the scsi#1's entry from fstab
3. cretaed dirs for user home and web home in #2, and linked the original
names to the new dirs, and unzipped backups.
4. I restarted the box.
Here is what I find problems now:
5. Mail to virtual host gives error.
any mail to USER@VIRTHOST1.com being sent gives this error
554 MX list for VIRTHOST1.com. points back to MAINDOMAIN.com
554 <USER@VIRTHOST1.com>... Local configuration error
The MX record has not been updated recently, and before the disk crash
things were just great.
6. I did restart the box, and I can reach the web sites of the virtual
domains. I can nslookup them all. And the box is responding to all IPs.
Please see the extract from logs at the end.
Questions and Need-Help areas:
Q1. Should the MX record of a virtual host point to itself or to
the primary?
Q2. Since that updation takes time to take effect, is there
anything in sendmail I can configure to accept mails for the
virtual hosts as a quick fix?
Q3. Is qmail any better than sendmail in this regard?
Q4. After the device da1 is ignored for not finding the disk,
why does sendmail give error about fill_fd?
I have created links to the web and user homes, why should any software
care for the actual device where the names are mounted/linked?
Q5. How can I find remotely if the disk failed or the adapter cable or
something? I mean I bought the three SCSI disks same time, same place.
Any help will be highly appreciated. Please respond to shashi@WEBSI.com
since that is the only email address that seems to working currently.
Thanks in Advance,
Shashi
======== Start of Critical.log extract =========
Oct 17 09:22:25 WEBSI /kernel: npx0: INT 16 interface
Oct 17 09:22:25 WEBSI /kernel: Waiting 15 seconds for SCSI devices to settle
Oct 17 09:22:25 WEBSI /kernel: Unexpected busfree. LASTPHASE == 0xe0
Oct 17 09:22:26 WEBSI /kernel: SEQADDR == 0x153
Oct 17 09:22:26 WEBSI /kernel: Unexpected busfree. LASTPHASE == 0xe0
Oct 17 09:22:26 WEBSI /kernel: SEQADDR == 0x153
Oct 17 09:22:26 WEBSI /kernel: changing root device to da0s1a
Oct 17 09:22:26 WEBSI /kernel: da0 at ahc0 bus 0 target 0 lun 0
Oct 17 09:22:26 WEBSI /kernel: da0: <SEAGATE ST39140W 1487> Fixed Direct Access SCSI-2 device
Oct 17 09:22:26 WEBSI /kernel: da0: 40.000MB/s transfers (20.000MHz, offset 8, 16bit), Tagged Qu
eueing Enabled
Oct 17 09:22:26 WEBSI /kernel: da0: 8683MB (17783240 512 byte sectors: 255H 63S/T 1106C)
Oct 17 09:22:26 WEBSI /kernel: da2 at ahc0 bus 0 target 2 lun 0
Oct 17 09:22:26 WEBSI /kernel: da2: <SEAGATE ST39140W 1487> Fixed Direct Access SCSI-2 device
Oct 17 09:22:26 WEBSI /kernel: da2: 40.000MB/s transfers (20.000MHz, offset 8, 16bit), Tagged Qu
eueing Enabled
Oct 17 09:22:26 WEBSI /kernel: da2: 8683MB (17783240 512 byte sectors: 255H 63S/T 1106C)
Oct 17 09:22:26 WEBSI /kernel: cd0 at ahc0 bus 0 target 6 lun 0
Oct 17 09:22:26 WEBSI /kernel: cd0: <NEC CD-ROM DRIVE:465 1.03> Removable CD-ROM SCSI-2 device
Oct 17 09:22:26 WEBSI /kernel: cd0: 20.000MB/s transfers (20.000MHz, offset 15)
Oct 17 09:22:26 WEBSI /kernel: cd0: Attempt to query device size failed: NOT READY, Medium not p
resent
Oct 17 09:22:26 WEBSI /kernel: Unexpected busfree. LASTPHASE == 0xe0
Oct 17 09:22:26 WEBSI /kernel: SEQADDR == 0x153
Oct 17 09:22:26 WEBSI /kernel: Unexpected busfree. LASTPHASE == 0xe0
Oct 17 09:22:26 WEBSI /kernel: SEQADDR == 0x153
Oct 17 09:22:27 WEBSI /kernel: (da1:ahc0:0:1:0): got CAM status 0x53
Oct 17 09:22:27 WEBSI /kernel: (da1:ahc0:0:1:0): fatal error, failed to attach to device
Oct 17 09:22:27 WEBSI /kernel: (da1:ahc0:0:1:0): lost device
Oct 17 09:22:27 WEBSI /kernel: (da1:ahc0:0:1:0): removing device entry
Oct 17 09:22:38 WEBSI sendmail[439]: NOQUEUE: SYSERR: putoutmsg (NO-HOST): error on output chann
el sending "451 fill_fd: before main() initmaps: fd 1 not open: Bad file descriptor": Input/outp
ut error
Oct 17 09:22:38 WEBSI sendmail[439]: NOQUEUE: SYSERR(root): fill_fd: before main() initmaps: fd
1 not open: Bad file descriptor
Oct 17 09:22:38 WEBSI sendmail[439]: NOQUEUE: SYSERR(root): fill_fd: before main() initmaps: fd
2 not open: Bad file descriptor
Oct 17 10:25:36 WEBSI sendmail[1107]: KAA01105: SYSERR(root): MX list for shift-f1.com. points b
ack to WEBSI.com
Oct 17 18:38:59 WEBSI sendmail[4085]: SAA04083: SYSERR(root): MX list for shift-f1.com. points b
ack to WEBSI.com
======== End of Critical.log extract =========
--
Shashi Joshi
_____________________________________________________________________
__o o__ o__ o__ o__ There's one
_ \<._ _.>/ _ _.>/ _ _.>/ _ _.>/ _ in every
(_)/ (_) (_) \(_) (_) \(_) (_) \(_) (_) \(_) crowd...
_____________________________________________________________________
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?19991017170959.A4993>