Linux-Misc Digest #189, Volume #27 Wed, 21 Feb 01 18:13:02 EST
Contents:
FTP scripting... ("Ren� Scheibe")
Re: Verify crontab, please ("Peter T. Breuer")
Re: Size of LINUX (Rolie Baldock)
Re: Zipping an entire installation ("Peter T. Breuer")
Re: Size of LINUX (Rolie Baldock)
Re: Verify crontab, please (Jean-David Beyer)
Re: DAMN! my printer isn't printing :((( (Reiner Griess)
Netscape problems with 2.4.0 kernel ? (was Netscape 4, Mozilla errs with ("Marcello
M. Pavan")
Re: Size of LINUX (Rolie Baldock)
Re: Size of LINUX (Rolie Baldock)
Re: FTP scripting... (Frank da Cruz)
Re: netscape lock file ("D. Stimits")
Re: Cannot umount, ZIP questions (Jim Chisholm)
Loopback devices for testing RAID and LVM
(=?iso-8859-1?Q?Jos=E9_Luis_Domingo_L=F3pez?=)
----------------------------------------------------------------------------
From: "Ren� Scheibe" <[EMAIL PROTECTED]>
Subject: FTP scripting...
Date: Wed, 21 Feb 2001 23:12:57 +0100
...can someone tell me if the normal ftp-client is scriptable???
How can I write a script for it.
Can you give me an example???
I want to login to a server and put a file on it.
Th@nx
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: Verify crontab, please
Date: Wed, 21 Feb 2001 22:26:20 GMT
Joe Knapp <[EMAIL PROTECTED]> wrote:
> Warning: I think you may have been hacked. I came across your
> message because I noticed the exact same problem on my
> RedHat 6.2 machine and typed in the string "*/5 * * * * /usr/sbin/init"
Yes, that's a full-fledged remote crack. Ask (politely) the sysadmin at the
universities that you got by reverse dns of those IPs to look into it,
after checking the bugtraq record.
I'll leave the rest of your message intact, though I'd normally edit!
It's important that people get to see the whole attack signature.
Peter
> into DejaNews and sure enough you noticed the same strange
> crontab entry.
> Here is the chronology:
> On Feb. 19, I noticed that the "named" daemon was not running. It turned out
> that the named executable was MISSING. The /var/log/messages file showed
> that something bad
> had happened:
> Feb 19 06:07:47 copperas inetd[2706]: finger/tcp: unknown service
> Feb 19 06:12:22 copperas inetd[2706]: /usr/sbin/nmbd (pid 2719): exit status
> 1
> Feb 19 06:12:23 copperas inetd[2706]: /usr/sbin/nmbd (pid 2720): exit status
> 1
> Feb 19 06:12:23 copperas inetd[2706]: /usr/sbin/nmbd (pid 2721): exit status
> 1
> .
> .
> .
> Feb 19 08:59:38 copperas inetd[2706]: /usr/sbin/nmbd (pid 3335): exit status
> 1
> Feb 19 08:59:38 copperas inetd[2706]: netbios-ns/udp server failing (looping
> or
> being flooded), service terminated for 10 min
> I reinstalled the bind package to recover named and everything appeared to
> be OK. But then it turned out my cron jobs weren't running. This was the
> output of crontab -l:
> # DO NOT EDIT THIS FILE - edit the master and reinstall.
> # (/cr0n installed on Mon Feb 19 06:07:47 2001)
> # (Cron version -- $Id: crontab.c,v 2.13 1994/01/17 03:20:37 vixie Exp $)
> */5 * * * * /usr/sbin/init
> So it looks like the crontab was reloaded from a file /cr0n at 6:07:47 that
> morning, the exact time as the finger entry in the messages file, right
> before things got hosed. Furthermore,
> the date on cron itself was:
> -rwxr-xr-x 1 root root 318004 Feb 19 06:07 /usr/sbin/init
> Urp! Someone modified init and set it up to run every five seconds!
> There was also an in.smb file with the same timestamp.
> Running strings on init yielded the following possibly suspicious output:
> WVSU
> ][^_
> uW9M
> 4.40.1.231
> 129.21.3.102
> 128.104.4.36
> .la.pid
> socket
> bind
> recvfrom
> %s %s %s
> aIf3YWfOhw.V.
> PONG
> *HELLO*
> Now, why would IP addresses be embedded in init?
> Running nslookup on the above IP addresses yields:
> crtntx1-ar4-001-231.dsl.gtei.net
> grace.isc.rit.edu
> sky3.engr.wisc.edu
> I am still looking into this but be warned it might be some kind of crack
> job. I haven't seen anything else so far, so if this thing is malicious it
> hasn't sprung yet. Perhaps some kind of DOS attack in the making?
> Joe
> ~
> "Robert Jones" <[EMAIL PROTECTED]> wrote in message
> news:[EMAIL PROTECTED]...
>> Because of some hardware problems involving my tape drive, I screwed
>> around with crontab quite a bit in the past few days. With everything
>> fixed, I decided to remove the extra entries in crontab. I was surprised
>> to find they're already gone, along with the one I wanted to keep.
>> That's no problem; it's just a matter of doing it again. What bothers me
>> is an entry that I don't remember seeing before. Could someone verify
>> this should be in RH6.0?
>>
>> [root@localhost rj]# crontab -l
>> # DO NOT EDIT THIS FILE - edit the master and reinstall.
>> # (/tmp/crontab.20475 installed on Tue Feb 13 06:00:08 2001)
>> # (Cron version -- $Id: crontab.c,v 2.13 1994/01/17 03:20:37 vixie Exp
>> $)
>> */5 * * * * /usr/sbin/init
>>
>> Looks like I'm running init every 12 seconds.
>>
>> Thanks
>> --
>> It is better to live rich than to die rich.
>> -- Samuel Johnson
>>
>> 6:06am up 2 days, 13:58, 2 users, load average: 0.10, 0.12, 0.07
>>
>>
------------------------------
From: [EMAIL PROTECTED] (Rolie Baldock)
Subject: Re: Size of LINUX
Date: Wed, 21 Feb 2001 22:26:04 GMT
Hello Peter,
You might as well talk to me in "Ancient Swahili", I haven't a clue
what you are talking about. I gather that perhaps I should do nothing.
Leave LINUX to those who talk in "Ancient Swahili".
Regards,
On Tue, 20 Feb 2001 23:53:49 +0100, "Peter T. Breuer"
<[EMAIL PROTECTED]> wrote:
>Rolie Baldock <[EMAIL PROTECTED]> wrote:
>> Speed has never been my problem, and will be most unlikely problem in
>
>Speed IS your problem. That you are not concerned about it is another
>matter.
>
>> the future since it hasn't shown up already. My system is stable and
>> from the experiences of friends who are always BROKE because they are
>> always upgrading with little or no demonstrable benefit, I cannot see
>> a valid reason to consider it. My two servers are 33MHz the
>
>Well, a 33MHz 486 will not be able to support loads from several nfs
>clients at once. Try receiving a mail storm on an nfs mounted mail
>spool! My P450 is sometimes bogged down by 250 copies of procmail
>battling for a single lock on an nfs directory .. it can take hours
>to clear the backlogs of thousands of mail messages and error messages
>and rebound messages, etc ..
>
>Actually, a slow machine is actually useful as primary MX (my primary
>is a P100). It's self-throttling!
>
>> work-stations are 66MHz and I am a RETIRED P.E.E. I take life
>> leisurely and try not to give myself a stress problem. My original
>> question was and still is: can I change my two servers to LINUX
>> without a tremendous hardware upgrade? And if so how is it done?
>
>Just change them. No hardware is necessary. But you are severely
>limited by the 33MHz 486 cpu. It will be like wading through mud (I
>speak as one who owns a 16MHz 386sx and a 50MHz 486sx). It will
>work as an nfs file server, but you have to be careful to strip it down
>to just that. My 486 can handle X, but I wouldn't dream of trying more
>than a single shell plus an inetd and a portmapper on the 386! RIP.
>Do a basic install of debian or slackware, and then start hacking lumps
>off.
>
>Peter
--Rolie Baldock. email: <[EMAIL PROTECTED]>
Subtract one thousand and nine for direct email
------------------------------
From: "Peter T. Breuer" <[EMAIL PROTECTED]>
Subject: Re: Zipping an entire installation
Date: Wed, 21 Feb 2001 23:13:35 +0100
[EMAIL PROTECTED] wrote:
> whole thing and have a clean system again in a few minutes instead of
> hours. I tried going to the root directory and tar/bzipping the whole
> thing, but that didn't work. I don't have a CD-R, so this seems like the
In what way does it "not work"? It created a tar, did it not? And this
tar untars to give you back your fs state. So what else can you want!
> best way to back it all up. Any suggestions?
Use tar.
Or cpio. Or afio. Or just "dd" (over ssh, like me).
Peter
------------------------------
From: [EMAIL PROTECTED] (Rolie Baldock)
Subject: Re: Size of LINUX
Date: Wed, 21 Feb 2001 22:31:16 GMT
Hello Grant,
Do you talk "Ancient Swahili" like Peter? Well I haven't a clue what
he's talking about with the "nfs" and "mail storm". Since my LAN works
OK as is perhaps I should forget LINUX. That is the feeling I get from
these messages.
Regards,
On Wed, 21 Feb 2001 00:57:47 GMT, [EMAIL PROTECTED] (Grant Edwards)
wrote:
>In article <[EMAIL PROTECTED]>, Peter T. Breuer wrote:
>
>>Well, a 33MHz 486 will not be able to support loads from
>>several nfs clients at once. Try receiving a mail storm on an
>>nfs mounted mail spool! My P450 is sometimes bogged down by 250
>>copies of procmail battling for a single lock on an nfs
>>directory. .. it can take hours to clear the backlogs of
>>thousands of mail messages and error messages and rebound
>>messages, etc ..
>
>Mailservers and NFS: a combination almost gauranteed to be a
>disaster. :) Switching to maildir mailboxes might help...
>
>While running Linux on a 485-33 with 8/50 is an interesting
>exercise, if you really need the box to do useful work,
>upgrading is probably worth the effort.
>
>--
>Grant Edwards grante Yow! Let's all show human
> at CONCERN for REVEREND MOON's
> visi.com legal difficulties!!
--Rolie Baldock. email: <[EMAIL PROTECTED]>
Subtract one thousand and nine for direct email
------------------------------
From: Jean-David Beyer <[EMAIL PROTECTED]>
Subject: Re: Verify crontab, please
Date: Wed, 21 Feb 2001 17:34:25 -0500
Joe Knapp wrote (in part):
>
> Warning: I think you may have been hacked. I came across your
> message because I noticed the exact same problem on my
> RedHat 6.2 machine and typed in the string "*/5 * * * * /usr/sbin/init"
> into DejaNews and sure enough you noticed the same strange
> crontab entry.
>
> -rwxr-xr-x 1 root root 318004 Feb 19 06:07 /usr/sbin/init
Mine is not even there:
valinux:jdbeyer[~]$ ls -l /sbin/init
-rwxr-xr-x 1 root root 26660 Aug 1 2000 /sbin/init
valinux:jdbeyer[~]$
>
> Urp! Someone modified init and set it up to run every five seconds!
It is not as bad as that: it is only every five minutes. But that is
not very good either. But they had to be root already to write in
/sbin or /usr/sbin.
>
> Running strings on init yielded the following possibly suspicious output:
>
> WVSU
> ][^_
> uW9M
> 4.40.1.231
> 129.21.3.102
> 128.104.4.36
> .la.pid
> socket
> bind
> recvfrom
> %s %s %s
> aIf3YWfOhw.V.
> PONG
> *HELLO*
>
> Now, why would IP addresses be embedded in init?
Just to check, I ran strings on my init and observed no IP
addresses, nor did the words socket or bind or recvfrom appear. I
got a lot more output than you did, but you probably omitted the
boring stuff. My sysinit is from SysVinit-2.78-5.
--
.~. Jean-David Beyer Registered Linux User 85642.
/V\ Registered Machine 73926.
/( )\ Shrewsbury, New Jersey
^^-^^ 5:15pm up 1 day, 49 min, 3 users, load average: 2.10, 2.14,
2.06
------------------------------
From: [EMAIL PROTECTED] (Reiner Griess)
Crossposted-To: linux.debian.user
Subject: Re: DAMN! my printer isn't printing :(((
Date: Wed, 21 Feb 2001 22:00:36 +0100
Reply-To: [EMAIL PROTECTED]
On Mon, 19 Feb 2001 18:49:53 GMT,
Stephen Patterson <[EMAIL PROTECTED]> wrote:
>On Mon, 19 Feb 2001 16:09:14 +0100, Reiner Griess wrote:
>>Hi together,
>>
>>i have never had such problems with installing
>>a printer in RedHat. /etc/printcap is configured,
>>lpd is runnning (restarting does not help).
>>
>>The problem seems to be there with the parallel
>>port?! 'lpstat > /dev/lp0' should print
>>some ASCII chars to the printer - but nothing
>>happens! No errors in /var/log/messages...
>>The ASCIIs are sent to nirvana....
>>
>>Any ideas??
>>Thank you
>>
>>Reiner
>>
>>--
>/usr/sbin/magicfilterconfig --force is used to set up printers on debian.
>
>Apart from that, can you say what printer you are using (information is
>helpful)
Thanks, but magicfilter (--force??) hasn't helped.
I'd configured the printer that way... Will see
if the other replies will help...
bye
reiner
--
------------------------------
From: "Marcello M. Pavan" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.networking,comp.os.linux.x
Subject: Netscape problems with 2.4.0 kernel ? (was Netscape 4, Mozilla errs with
Date: Wed, 21 Feb 2001 14:35:24 -0800
Reply-To: [EMAIL PROTECTED]
Greetings all,
I recently posted a problem I had with NS4.76 not connecting
to various common web sites, and it truns out that the problem is
the 2.4.0 kernel. The problem surfaced also with Mozilla and Lynx, so it
had to be bigger than NS.
I compiled and am now running the 2.2.19pre14 kernel,
and now it works fine, just as before the RH7 upgrade.
anyone know why this happened ? are there patches for 2.4.0
to fix this ?
regards,
BuN
------------------------------
From: [EMAIL PROTECTED] (Rolie Baldock)
Subject: Re: Size of LINUX
Date: Wed, 21 Feb 2001 22:37:03 GMT
OK...OK... I have a few spare 127 Mb IDE drives and there are a few
more down at the shop and I can find a couple of 16Mb simms to insert
in the servers. But is is so difficult to get LINUX to work as a file
server ONLY with this hardware set up? Moreover is it worth the
trouble? I have dedicated DX2-66s (plenty of them) for dedicated jobs
so I ONLY need the servers to do that, NOTHING ELSE.
On Tue, 20 Feb 2001 23:03:54 GMT, [EMAIL PROTECTED] (Grant Edwards)
wrote:
>In article <[EMAIL PROTECTED]>, Rolie Baldock wrote:
>
>>Speed has never been my problem, and will be most unlikely problem in
>>the future since it hasn't shown up already. My system is stable and
>>from the experiences of friends who are always BROKE because they are
>>always upgrading with little or no demonstrable benefit, I cannot see
>>a valid reason to consider it. My two servers are 33MHz the
>>work-stations are 66MHz and I am a RETIRED P.E.E. I take life
>>leisurely and try not to give myself a stress problem. My original
>>question was and still is: can I change my two servers to LINUX
>>without a tremendous hardware upgrade? And if so how is it done?
>
>8M is enough RAM to run Linux (with no graphic stuff). But,
>most "standard" distibutions won't install with less than 16M
>(debian and slackware might).
>
>50MB is barely enough disk space to Linux. You can do it, but
>it's tough. You really need to know what you're doing to get
>Linux trimmed down that small. I've had Linux running in 80MB,
>and that was tight.
>
>There are various "tiny Linux" distributions that might work
>for you:
>
>http://tiny.seul.org/en/
>http://www.txdirect.net/users/mdfranz/tinux.html
>
>--
>Grant Edwards grante Yow! "THE LITTLE PINK
> at FLESH SISTERS," I saw them
> visi.com at th' FLUROESCENT BULB
> MAKERS CONVENTION...
--Rolie Baldock. email: <[EMAIL PROTECTED]>
Subtract one thousand and nine for direct email
------------------------------
From: [EMAIL PROTECTED] (Rolie Baldock)
Subject: Re: Size of LINUX
Date: Wed, 21 Feb 2001 22:41:21 GMT
Sorry Peter,
Like I said before I simply do not understand your message.
What you mean by 2507 /bin
and 3460 /boot
means absolutely nothing to me.
On Wed, 21 Feb 2001 01:04:07 +0100, "Peter T. Breuer"
<[EMAIL PROTECTED]> wrote:
>Grant Edwards <[EMAIL PROTECTED]> wrote:
>> In article <[EMAIL PROTECTED]>, Rolie Baldock wrote:
>
>>>Speed has never been my problem, and will be most unlikely problem in
>>>the future since it hasn't shown up already. My system is stable and
>>>from the experiences of friends who are always BROKE because they are
>>>always upgrading with little or no demonstrable benefit, I cannot see
>>>a valid reason to consider it. My two servers are 33MHz the
>>>work-stations are 66MHz and I am a RETIRED P.E.E. I take life
>>>leisurely and try not to give myself a stress problem. My original
>>>question was and still is: can I change my two servers to LINUX
>>>without a tremendous hardware upgrade? And if so how is it done?
>
>> 8M is enough RAM to run Linux (with no graphic stuff). But,
>> most "standard" distibutions won't install with less than 16M
>> (debian and slackware might).
>
>> 50MB is barely enough disk space to Linux. You can do it, but
>> it's tough. You really need to know what you're doing to get
>> Linux trimmed down that small. I've had Linux running in 80MB,
>> and that was tight.
>
>Debian is easily cuttable to 8MB. My / partition on standard debians
>runs about 26MB, including several kernels and modules. Just hack that
>(or a slackware) down a bit.
>
>2507 /bin
>3460 /boot
>26 /dev
>8002 /etc
>1 /floppy
>201 /home
>19663 /lib
>2452 /sbin
>
>Peter
--Rolie Baldock. email: <[EMAIL PROTECTED]>
Subtract one thousand and nine for direct email
------------------------------
From: [EMAIL PROTECTED] (Frank da Cruz)
Subject: Re: FTP scripting...
Date: 21 Feb 2001 22:51:50 GMT
In article <971eed$n6ig6$[EMAIL PROTECTED]>,
Ren� Scheibe <[EMAIL PROTECTED]> wrote:
: ...can someone tell me if the normal ftp-client is scriptable???
:
It is not scriptable.
: How can I write a script for it.
:
You can use a different FTP client that IS scriptable:
http://www.columbia.edu/kermit/ftpclient.html
: Can you give me an example???
:
Yes:
http://www.columbia.edu/kermit/ftpscript.html
- Frank
------------------------------
Date: Wed, 21 Feb 2001 15:52:49 -0700
From: "D. Stimits" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: netscape lock file
Jeff Kempe wrote:
>
> James Silverton wrote:
>
> > Eric Wertman wrote:
> > >
> > > All-
> > >
> > > Netscape continues to complain about a $HOME/.netscape/lock file, when
> > > no such file exists. Anyone seen this and know the answer? Thanks!
> > >
> > > Eric Wertman
> >
> > In my experience, it is usually caused by attempting to have more than
> > one instance of Netscape open at the same time. I have even managed it
> > by accidentally double-clicking the Netscape item in KDE!
> >
> > Jim
> > --
> > James V. Silverton
> > Potomac, Maryland.
>
> What version of Netscape are you using?
>
> I checked my own ~/.netscape/ (with ls -al) and the 'lock' link was there,
> although it was broken (pointed to
> an invalid socket). It doesn't seem to matter to my installation whether
> that link is there or not. I can start as
> many netscape sessions as I want.
The link is just a kind of marker, it is intentionally not pointing at
anything real. I would suspect that if the link itself does not exist,
and it complains of a lock, possibly there is a permissions problem
where netscape can't access the directory. Try moving the dir itself to
a backup as a test, and let netscape recreate the directory. From ~, do
"mv .netscape .netscape-backup" or something similar. If this fixes it,
you can move it back, and try working with chmod. Before doing any of
this, try killing netscape, do a "killall -9 netscape" and "killall -9
netscape-communicator" (it might be running under either of those
names).
>
> My advice would be to get the latest distribution of Netscape (4.* ) and
> reinstall it.
>
> --
> Jeff Kempe
> Research Programmer
> Physics Research and Education
> London Regional Cancer Center
> 790 Commissioners Road East
> London, Ontario
> N6A 4L6
>
> Office: (519) 685-8600 ext. 53318
> Fax: (519) 685-8658
>
> -----BEGIN PGP PUBLIC KEY BLOCK-----
> Version: GnuPG v1.0.1 (GNU/Linux)
> Comment: For info see http://www.gnupg.org
>
> mQGiBDp3CyYRBADSiPk87S23lfGp6gBuDmpWQXOl6oXpnFAzUFxIrRymIFgx9L0B
> U95AOAm9GyX5aT+XeEX9vjxrjZzfzXBMizLpswuaz4fg6nLNh4bq2q0NfEioON9O
> DLDK8vq3rYz4/jLt9NBQFRF/X6O3AkBsrOzSATMeS8IgW+qYpk7DZy6hQwCgsYNF
> 5sG+0s4QY6N+OgkX+iENxysEAJTVQzCwxCsN4dv60eNuIsGkuKI6ppraTwhEaghy
> HnirYe5AgNXeopy1aEQaPQw5MaTUWirE1xxlR++0ZWf4j1iq2r0jMOQ4uO0ZywLX
> DYgjwNJ5uq39a17tH+eTMQPGu3L9/2U7g3IH+C9qRY6T3fjr5HT81bGDjByTJ+A7
> P3EPBAC+53Ht0Uq2UeeFil2+4UKKg75GbX8ip9YTQBy4ANnBnxZ0mZFvOpfFxPU9
> 56vQOcfY+5gM6dzSsn04Uuja3Jhvt7HQ0yOjMCi0YqiKehiZXdn/s+9/zgLGbw/I
> ndhSXb8hB8yPQhI9geMFf5DyH2Rgp1ueD8eNm5YSw9X3SMSETLRBSmVmZnJleSBL
> ZW1wZSAoTFJDQyBQaHlzaWNzIENvbXB1dGVyIEd1eSkgPGprZW1wZUBwaHkubHJj
> Yy5vbi5jYT6IVgQTEQIAFgUCOncLJgQLCgQDAxUDAgMWAgECF4AACgkQT
------------------------------
From: Jim Chisholm <[EMAIL PROTECTED]>
Subject: Re: Cannot umount, ZIP questions
Date: Wed, 21 Feb 2001 18:58:07 -0400
==============CF36FC6499AD691948D5DF89
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
The Spook wrote:
> Gary wrote ...
> >
> >I'm using RH 7.0, and after mounting a CDROM, or ZIP drive,
> >I cannot
> >unmount it or eject the media without doing a shutdown. I
> >tried the -f
> >switch without luck, does anyone have a suggestion on how to
> >force
> >an unmount? Perhaps something like unmount /dev/cdrom
> >-RIGHT_NOW :-)
> >
> >Also, depending on the file system on parallel ZIP disk, my
> >/dev/sda# is
> >different, sda1, or sda4. This makes it necessary to know
> >the
> >filesystem of the disk before mounting, OR try mount
> >/dev/sda1
> >and mount /dev/sda4. I "thought" that the device assignments
> >were
> >physical, not logical, this appears to not be the case am I
> >missing
> >something?
> >
> >
> >
> > Regards,
> >
> > Gary
>
> You cannot unmount a filesystem (i.e. a CD or ZIP-disk here) if there are
> files (or directories) open on the filesystem. The open files may be due to
> running programs (daemons, background jobs, ...) or to a shell (OK, that's a
> running program too) with its current directory in the filesystem.
>
> To find all programs with open files with the corresponding open files (and
> directories) on a filesystem mounted as (say) /mnt/cdrom, use the command:
>
> lsof +d /mnt/cdrom
>
> Look for the offending program(s) listed in the first column of the output
> (truncated to nine characters on my system) -- kill them or cd to another
> directory not in the filesystem (for shells).
>
> /TRY
fuser -km /dev/cdrom is a sure-fire way to free it up..
Jim
--
=======================================================
Jim Chisholm
Dalhousie University, Dept. Physics Halifax N.S. Canada
Halifax Regional Fire and Emergency Service
Captain/Training Officer Bay Road Station 59
=======================================================
==============CF36FC6499AD691948D5DF89
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
The Spook wrote:
<blockquote TYPE=CITE>Gary wrote ...
<br>>
<br>>I'm using RH 7.0, and after mounting a CDROM, or ZIP drive,
<br>>I cannot
<br>>unmount it or eject the media without doing a shutdown. I
<br>>tried the -f
<br>>switch without luck, does anyone have a suggestion on how to
<br>>force
<br>>an unmount? Perhaps something like unmount /dev/cdrom
<br>>-RIGHT_NOW :-)
<br>>
<br>>Also, depending on the file system on parallel ZIP disk, my
<br>>/dev/sda# is
<br>>different, sda1, or sda4. This makes it necessary to know
<br>>the
<br>>filesystem of the disk before mounting, OR try mount
<br>>/dev/sda1
<br>>and mount /dev/sda4. I "thought" that the device assignments
<br>>were
<br>>physical, not logical, this appears to not be the case am I
<br>>missing
<br>>something?
<br>>
<br>>
<br>>
<br>> Regards,
<br>>
<br>> Gary
<p>You cannot unmount a filesystem (i.e. a CD or ZIP-disk here) if there
are
<br>files (or directories) open on the filesystem. The open files may be
due to
<br>running programs (daemons, background jobs, ...) or to a shell (OK,
that's a
<br>running program too) with its current directory in the filesystem.
<p>To find all programs with open files with the corresponding open files
(and
<br>directories) on a filesystem mounted as (say) /mnt/cdrom, use the command:
<p> lsof +d /mnt/cdrom
<p>Look for the offending program(s) listed in the first column of the
output
<br>(truncated to nine characters on my system) -- kill them or cd to another
<br>directory not in the filesystem (for shells).
<p> /TRY</blockquote>
fuser -km /dev/cdrom is a sure-fire way to free it up..
<p>Jim
<pre>--
=======================================================
Jim
Chisholm <[EMAIL PROTECTED]>
Dalhousie University, Dept. Physics Halifax N.S. Canada
Halifax Regional Fire and Emergency Service
Captain/Training Officer Bay Road Station 59
=======================================================</pre>
</html>
==============CF36FC6499AD691948D5DF89==
------------------------------
From: [EMAIL PROTECTED] (=?iso-8859-1?Q?Jos=E9_Luis_Domingo_L=F3pez?=)
Subject: Loopback devices for testing RAID and LVM
Date: Thu, 22 Feb 2001 00:01:53 +0000
Reply-To: [EMAIL PROTECTED]
=====BEGIN PGP SIGNED MESSAGE=====
Hash: SHA1
Hi people:
I would like to give Linux software RAID and LVM implementations a try,
but I don't have spare hardware to make "real" test. I've heard there is a
way to simulate real devices using the so called loopback devices (I
already use them to mount ISO images before burning CD).
But when I try something similar to create fake devices (dd if=/dev/zero
of=/tmp/file0[1-4].bin), and configure a RAID array with them, mkraid
complains about improper device types.
So, how can I make RAID and/or LVM volumes using loopback devices ?. Seems
that I'm quite near, but not there still :)
- --
Jos� Luis Domingo L�pez
Linux Registered User #189436 Debian GNU/Linux Potato (P166 64 MB RAM)
jdomingo EN internautas PUNTO org => � Spam ? Atente a las consecuencias
jdomingo AT internautas DOT org => Spam at your own risk
=====BEGIN PGP SIGNATURE=====
Version: GnuPG v1.0.4 (GNU/Linux)
Comment: For info see http://www.gnupg.org
iEYEARECAAYFAjqUVu8ACgkQao1/w/yPYI3tUgCeNrzQ0bxVixwSzIFZHfdP65uV
x0AAn3cNspEAEOZVHEwMpmwooLv6oLr3
=v+Ii
=====END PGP SIGNATURE=====
------------------------------
** FOR YOUR REFERENCE **
The service address, to which questions about the list itself and requests
to be added to or deleted from it should be directed, is:
Internet: [EMAIL PROTECTED]
You can send mail to the entire list by posting to comp.os.linux.misc.
Linux may be obtained via one of these FTP sites:
ftp.funet.fi pub/Linux
tsx-11.mit.edu pub/linux
sunsite.unc.edu pub/Linux
End of Linux-Misc Digest
******************************
