Linux-Misc Digest #817, Volume #26 Sun, 14 Jan 01 21:13:02 EST
Contents:
Re: Mandrake PATH setting (* Tong *)
Re: Absolute File Size (Dances With Crows)
Re: destination host unreachable (Dances With Crows)
Re: fetchmail (Brad Bailey)
Re: Which prog for streaming (Michael Heiming)
Help, newbie cannot boot up RH 6.2 ("Jon Cook")
Re: file permissions suddenly changed!!!!!! (Harry Putnam)
Re: bad superblock but no problem mounting? (E J)
ipchains log entry -- meaning? (Luben Tuikov)
Re: fetchmail (Bob Schreibmaier)
Monitor Processes ("Brian E. Seppanen")
----------------------------------------------------------------------------
From: * Tong * <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux.mandrake
Subject: Re: Mandrake PATH setting
Date: 14 Jan 2001 20:32:56 -0400
[EMAIL PROTECTED] (Bit Twister) writes:
> On 13 Jan 2001 17:06:24 -0400, * Tong * <[EMAIL PROTECTED]> wrote:
> >Hi,
> >
> >I choose high security level when installing mandrake and found that
> >my PATH setting in /etc/profile no longer work any more (at least
> >for the root). Instead, the comment hints that the Mandrake security
> >management is taking care of it.
> >
>
> I have created /etc/profile.d/xx_local.sh
> with my changes and did a
> chmod +x /etc/profile.d/xx_local.sh
> I am running at SECURE_LEVEL=3
Can someone explain this a little bit? I have no clue how I should
do. Does the SECURE_LEVEL has anything to do with the xx_?
Thanks
--
Tong (remove underscore(s) to reply)
http://members.xoom.com/suntong001/
- All free contribution & collection & music from the heavens
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: Absolute File Size
Date: 15 Jan 2001 00:44:08 GMT
Reply-To: [EMAIL PROTECTED]
On 14 Jan 2001 22:24:19 GMT, Jim Pogras staggered into the Black Sun and said:
>I use a 30gig IDE drive to back up my Linux partitions and the Windows
>drives on my home network. I have upgraded to V2.4.0 and now I can only
>create a file with a max size of 107780391 bytes. Has anyone seen this?
>I've looked throught the Linux Docs BUT have not had any luck.
?? That's only 102 megabytes, and is a serious cause for concern. Or
did you misspeak, and mean "107780391 K" which is much larger than the
max capacity of the drive? Is there enough free space on the disk?
Which distro are you using? RedHat 7.0 at least shipped with a patched
2.2 kernel and a patched libc to handle files larger than 2G on 32-bit
systems. If you upgraded the kernel to 2.4.0 without recompiling libc
against 2.4.0 kernel headers, things could go wrong in various ways.
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: destination host unreachable
Date: 15 Jan 2001 00:44:10 GMT
Reply-To: [EMAIL PROTECTED]
On Sun, 14 Jan 2001 19:36:30 GMT, Steve Connet staggered into the Black
Sun and said:
>[EMAIL PROTECTED] (Dances With Crows) writes:
>> Weird. Looked in the system log (often in /var/log/messages or
>> /var/log/warn) for anything relating to networking problems? And what
>Yeah, nothing wierd in the /var/log/messages file. I don't have a
>/var/log/warn file. I would like however, if anything bad happens or
>suspicious happens that root is sent a email with the message. Is this
>possible?
man 5 syslog.conf for starters, though there doesn't seem to be any
built-in functionality for sending mail. I suppose you could have
something watching on a named pipe, but that's sort of overkill. SuSE
sets up their basic install to send messages to /dev/tty10 with this
line in /etc/syslog.conf :
kern.warn;*.err;authpriv.none /dev/tty10
so all the console operator has to do is press Ctrl-Alt-F10 to see
what's up.
>One is a DLink which uses the tulip.o driver, and the other I forgot
>but it uses the rhine.o driver. It is the one the cable company gave
>me when I got internet cable.
Which one is eth0 and which one is eth1? If eth1 is generating errors
as you mention a bit later, knowing which make+model it is could help.
>> As a hackaround/diagnostic, you could set up a cron job on the Linux box
>> that pings the 2K box every 5 minutes and spits out an error if it
>> doesn't ping successfully. Leave both machines on overnight, check in
>Yes I will try that, good idea. Now I gotta figure out how to setup a
>cron job. Never done it before... can you tell I'm new?
man 5 crontab and be prepared to get your brain dirty.
Or just execute the following script from the command line:
while ping -c1 win2Kbox ; do sleep 300 ; done ; echo -e "\007" ; date &
which will ping every 5 minutes, doing nothing until it can't ping, upon
which it will beep and echo the current date to the console. (Or you
could feed date's output to "mail"!)
>Also I was wondering if it could be some sort of routing problem? Like
>after a while when there is no traffic between the 2 machines maybe
>the routing table gets messed up or something? I don't know.
Very doubtful.
>eth1 Link encap:Ethernet HWaddr 00:A0:CC:24:F1:CD
> inet addr:192.168.1.1 Bcast:192.168.1.255 Mask:255.255.255.0
> UP BROADCAST RUNNING MULTICAST MTU:1500 Metric:1
> RX packets:0 errors:0 dropped:0 overruns:0 frame:0
> TX packets:0 errors:190 dropped:0 overruns:0 carrier:380
>
>It shows many TX errors. Here is my routing table using netstat -r:
>Destination Gateway Genmask Flags MSS Window irtt Iface
>192.168.1.0 * 255.255.255.0 U 0 0 0 eth1
>24.21.64.0 * 255.255.255.0 U 0 0 0 eth0
>127.0.0.0 * 255.0.0.0 U 0 0 0 lo
>default 24.21.64.1 0.0.0.0 UG 0 0 0 eth0
>Does it look suspicious at all?
Routing table looks fine. The number of errors in eth1 makes me
suspicious. Kernel version, distro, and make+model of eth1 might help
someone to find out if there's a hardware problem, or a kernel module
problem, or what.
--
Matt G|There is no Darkness in Eternity/But only Light too dim for us to see
Brainbench MVP for Linux Admin / Workin' in a code mine, hittin' Ctrl-Alt
http://www.brainbench.com / Workin' in a code mine, whoops!
=============================/ I hit a seg fault....
------------------------------
From: [EMAIL PROTECTED] (Brad Bailey)
Subject: Re: fetchmail
Reply-To: [EMAIL PROTECTED]
Date: Mon, 15 Jan 2001 00:45:59 GMT
In <[EMAIL PROTECTED]>, Steve Connet
<[EMAIL PROTECTED]> wrote:
:When I run fetchmail as user blix it works great. User blix as a
:~/.fetchmailrc file.
:
:But I want fetchmail to automatically start when I boot up the
:machine. If I put fetchmail in my rc.local will it run as root? And
:won't that look for a .fetchmailrc for root? And if I tell fetchmail
:where blix's .fetchmailrc file is, won't it send all mail fetched to
:root?
:
:How do I get fetchmail to start on bootup and fetchmail for user
:'blix'?
Put it in user blix's crontab, and run it a couple of times per hour.
It will run as user blix, and will be restarted if something goes wrong
with the fetchmail process.
Regards, Brad.
--
"If you have any trouble sounding condescending, find a Unix user to
show you how it's done." --Scott Adams
------------------------------
Date: Mon, 15 Jan 2001 01:41:54 +0100
From: Michael Heiming <[EMAIL PROTECTED]>
Subject: Re: Which prog for streaming
Hello,
[EMAIL PROTECTED] wrote:
> In article <[EMAIL PROTECTED]>,
> Michael Heiming <[EMAIL PROTECTED]> wrote:
> > Hello,
>
> >
> > I have no problem keeping the drive (ex. HP SureStore DLT 40) moving
> > all the time, I don't understand what you mean?
>
> That particular one is relatively slow. I was thinking in the 20+
> MB/s range. You may still be getting performance loss. Only way to
> tell is time a large backup figure out your average throughput. That
> drive should get 1.5 to 3MB/s at full speed depending on the
> compression rate.
Must be pretts expensive those 20+ MB/s, you're right about the speed the
HP
makes, I thought it would be fast, considering that it's not really
cheap.
>
>
> > If you had read the docs you would have read:
> >
> > --label
> >
> > which you can use/extend in a script with something like this:
> >
> > "Level-${LEVEL} Backup ${NOW} from ${HOSTNAME}"
> >
> > which gives you perfect control of your tape. As for labeling, it
> shouldn't
> > be to hard to use a pencil and write the same as
> > --label gives you on those pice of thick paper that come with new
> tapes.
>
> Not quite what I meant. Say you have a jukebox with 50 tapes, do you
> really want to have to remember which tape is in which slot? The
> backup software uses bar code labels and electronic labels on the
> media to track what's where and automatically build a new inventory
> when you swap tapes out. You can't track tapes by what backup is on
> them since they may contain many backups or none at all. You also
> want to track things like number of mounts, so you can retire tapes as
> they age.
>
> >
> > Scheduling is not so complicated, make once a plan, considering what
> > you need to backup how often.
>
> I does get to be a real pain when you have a lot of systems. You
> probably want to rotate full backups and incremental backups over your
> systems, so they're not all dumping all their data on any one night
> and clogging up the network. Then if one bombs out one night, you
> have to do a one-off to get it back to date. A good system will let
> you set the schedule and then never think about it again. There's
> also the issue of what tapes need to be shipped offsite and when they
> should be returned. Gets even worse when you have to work with
> database down-time windows.
>
> >
> > Reliability?
> >
> > Wow, AFAIK tar is used for backups since decades, I don't think that
> there is
> > any other solution as reliable as tar!
> >
> > All the comercial solutions I used/tested were just a pice of crap
> > and mostly unsecure as hell...
>
> I did an extensive system with Netbackup. A few terabytes of data and
> a 15,000 slot robot. Wasn't the easiest to get configured but hummed
> right along with rarely a problem once it got going. I've used Legato
> Networker on a number of smaller systems and it's a snap to get going.
> It all cases, they were using secure networks, so the IP based
> authentication was good enough. In a riskier environment, you'd have
> to add key pair encryption.
>
> Anyway, this is really comparing apples and oranges. Tar is an
> archiving tool and not a backup system. Netbackup actually stores
> it's files in tar format. If you want to use just tar, then you'll
> need a human as the backup system.
>
> Sent via Deja.com
> http://www.deja.com/
You quite right with backups on large systems where you have
robots/jukeboxes that change tapes automagically.
But from the original poster it doesn't sound as if he has such
enviroment, with only a couple of server you can
easily go the way I described, as a human has always to change the
tapes....:-(
Michael Heiming
------------------------------
From: "Jon Cook" <[EMAIL PROTECTED]>
Subject: Help, newbie cannot boot up RH 6.2
Date: Sun, 14 Jan 2001 20:20:33 -0500
I just installed Redhat 6.2 on a new machine and it will not boot. Lilo
comes up and starts to load linux, but then it halts and spits out this
message:
killing interrupt handler >Kernel panic: Attempted to
kill the idle task >In swapper task - not syncing
I had this happen when I tried to initially install booting from cd, but
then I downloaded boot.img from redhat lastnight and got the install to
work.
PC is stripped, only trident video card, kb and mouse.
AMD 1000 Athlon, Abit MB, IBM 46GB hd, 128MB RAM, Floppy, IDE Cdrom drive.
Any ideas? suggestions? Thanks in advance
Jon Cook
[EMAIL PROTECTED]
------------------------------
From: Harry Putnam <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.admin,comp.os.linux.help,comp.os.linux.security
Subject: Re: file permissions suddenly changed!!!!!!
Date: 14 Jan 2001 14:24:39 -0800
Jan Johansson <[EMAIL PROTECTED]> writes:
> >chmod -R 000 /lib would do that...
>
> Uhm, but that wouldtn crop the files to 0 bytes :)
Any one noticed the fishy date `Jan 1 1970 ' Isn't that the epochal
date that unix time is calulated from.
So we've got zero permissions, zero length files, all timestamped at
unix epochal time.
Could a major power break while curses was in use do something like
this? ... naa
------------------------------
From: E J <[EMAIL PROTECTED]>
Subject: Re: bad superblock but no problem mounting?
Date: Mon, 15 Jan 2001 01:32:51 GMT
Try the following:
# e2fsck /dev/hdb1
Mat wrote:
> even though hdb is giving me no problems i went to run e2fsck the other
> day and got the following
>
> <output>
> [root@weyland /root]# e2fsck /dev/hdb
> e2fsck 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
> Group descriptors look bad... trying backup blocks...
> e2fsck: Bad magic number in super-block while trying to open /dev/hdb
>
> The superblock could not be read or does not describe a correct ext2
> filesystem. If the device is valid and it really contains an ext2
> filesystem (and not swap or ufs or something else), then the superblock
> is corrupt, and you might try running e2fsck with an alternate
> superblock:
> e2fsck -b 8193 <device>
> </output>
>
> *********************
>
> tried it on /dev/hda and got the same, but for now am only talking about
> /dev/hdb
>
> to check my blocksize before using e2fsck -b i ran dumpe2fs and got the
> following...
>
> <output>
> [root@weyland /root]# dumpe2fs /dev/hdb
> dumpe2fs 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
> Filesystem volume name: <none>
> Last mounted on: <not available>
> Filesystem UUID: b03d299a-5a69-11d4-9f12-e51180f8c152
> Filesystem magic number: 0xEF53
> Filesystem revision #: 0 (original)
> Filesystem features: (none)
> Filesystem state: clean
> Errors behavior: Continue
> Filesystem OS type: Linux
> Inode count: 5017600
> Block count: 20066251
> Reserved block count: 1003312
> Free blocks: 19243037
> Free inodes: 5017589
> First block: 1
> Block size: 1024
> Fragment size: 1024
> Blocks per group: 8192
> Fragments per group: 8192
> Inodes per group: 2048
> Inode blocks per group: 256
> Last mount time: Sat Jul 15 16:56:15 2000
> Last write time: Sat Jul 15 16:56:41 2000
> Mount count: 3
> Maximum mount count: 20
> Last checked: Sun Jul 16 02:29:20 2000
> Check interval: 15552000 (6 months)
> Next check after: Fri Jan 12 03:29:20 2001
> Reserved blocks uid: 0 (user root)
> Reserved blocks gid: 0 (group root)
> ext2fs_read_bb_inode: Illegal indirect block found
> </output>
>
> and don't like the look of the last line :(
>
> *********************
>
> anyway, with the blocksize of 1024 i gave e2fsck another go with alt
> superblock
>
> <output>
> [root@weyland /root]# e2fsck -b 8193 /dev/hdb
> e2fsck 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
> e2fsck: Bad magic number in super-block while trying to open /dev/hdb
>
> The superblock could not be read or does not describe a correct ext2
> filesystem. If the device is valid and it really contains an ext2
> filesystem (and not swap or ufs or something else), then the superblock
> is corrupt, and you might try running e2fsck with an alternate
> superblock:
> e2fsck -b 8193 <device>
> </output>
>
> also tried alternate superblocks 16386, 24579, 32772... 65544 but i have
> the
> feeling this is not the problem
>
> *********************
>
> fdisk list gives
>
> <output>
> [root@weyland /root]# fdisk -l
>
> Disk /dev/hda: 128 heads, 63 sectors, 779 cylinders
> Units = cylinders of 8064 * 512 bytes
>
> Device Boot Start End Blocks Id System
> /dev/hda1 * 1 229 923296+ 83 Linux
> /dev/hda2 230 779 2217600 5 Extended
> /dev/hda5 230 737 2048224+ 83 Linux
> /dev/hda6 738 763 104800+ 83 Linux
> /dev/hda7 764 779 64480+ 82 Linux swap
>
> Disk /dev/hdb: 32 heads, 63 sectors, 19906 cylinders
> Units = cylinders of 2016 * 512 bytes
>
> Device Boot Start End Blocks Id System
> /dev/hdb1 1 19906 20065216+ 83 Linux
> </output>
>
> *********************
>
> does anyone have any clues as to what is wrong with my superblock(s)? as
> i said the drive is functioning ok apart from this....
>
> mat
------------------------------
From: Luben Tuikov <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.redhat
Subject: ipchains log entry -- meaning?
Date: Mon, 15 Jan 2001 01:34:47 GMT
RH7, 2.2.18
This is the /var/log/messages log entry I get when using
realplayer/msmediaplayer from an internal machine (masquaraded):
Jan 14 14:53:03 lt1 kernel: Packet log: input REJECT eth1 PROTO=2
24.43.3.1:65535 224.0.0.1:65535 L=32 S=0x00 I=43404 F=0x0000 T=1
O=0x00000494 (#5)
lt1 is my linux box, eth1 is the external if.
What does this exactly mean?
My guess is: input chain, on eth1 (my external NIC), guessing that proto
2 is TCP, .... and then is a mystery.
My default policy for the input chain is REJECT, and the reject rules
are:
ipchains -A input -i $extint -s $intnet -d 0.0.0.0/0 -l -j REJECT
and
ipchains -A input -s 0.0.0.0/0 -d 0.0.0.0/0 -l -j REJECT
I can't see how the first reject mathces so then it must be the
second...
So what does this log file mean and how can I fix this to work?
Module ip_masq_raudio is alredy in the kernel. I'm on a @home network so
it is possible that they are doing some tricks and redirecting realaudio
streams through a different machine, namely 24.43.3.1 ...)
TIA.
--
Luben
------------------------------
From: [EMAIL PROTECTED] (Bob Schreibmaier)
Subject: Re: fetchmail
Date: Mon, 15 Jan 2001 01:42:34 GMT
>In <[EMAIL PROTECTED]>, Steve Connet
><[EMAIL PROTECTED]> wrote:
>:When I run fetchmail as user blix it works great. User blix as a
>:~/.fetchmailrc file.
>:
>:But I want fetchmail to automatically start when I boot up the
>:machine. If I put fetchmail in my rc.local will it run as root? And
>:won't that look for a .fetchmailrc for root? And if I tell fetchmail
>:where blix's .fetchmailrc file is, won't it send all mail fetched to
>:root?
>:
>:How do I get fetchmail to start on bootup and fetchmail for user
>:'blix'?
On my system, I created a .fetchmailrc file in the root directory
and put the following line in the rc.local file:
/usr/local/bin/fetchmail -f /root/.fetchmailrc
The /root/.fetchmailrc file looks like this:
set daemon 120
poll mail.my-isp.net protocol pop3 no dns timeout 60:
user user1 pass user1password to user1 here fetchall
user user2 pass user2password to user2 here fetchall
This causes fetchmail to contact the mail server every two minutes
and download e-mail for user1 and user2 on my local system (they
have separate mailboxes on my ISP). Fetchmail times out if there
is no response from the mail server in 60 seconds. Of course,
user1password is user1's password on the mail server and user2password
is user2's password on the mail server.
I hope this is useful to you.
Bob
--
+------------------- \-\-\-\ ----------------------------+
| Bob Schreibmaier K3PH | E-mail: [EMAIL PROTECTED] |
| Kresgeville, PA 18333 | http://www.qsl.net/k3ph |
+--------------------------------------------------------+
------------------------------
From: "Brian E. Seppanen" <[EMAIL PROTECTED]>
Crossposted-To: comp.unix.admin
Subject: Monitor Processes
Date: Sun, 14 Jan 2001 20:34:02 -0500
What have others found to be a good way of monitoring processes on the
varies types of unices? Currently we do a grep for the process
information, but I'm wondering if there isn't something that might work
better. This would be on a variety of solaris 2.6, 2.7 and redhat
linux 6.2, 7.0 machines. Does the pid file get changed in any way to
indicate whether a process is still running?
Thanks,
Brian E. Seppanen
[EMAIL PROTECTED]
------------------------------
** 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
******************************
