md device failure (help!)
Hi all, my md array just crapped out on me! I'm partly responsible, since one of the device in the RAID5 array died some time ago and I neglected to replace it, but I don't think it's the whole problem now. When I assemble the array I get the following: r...@server:~# mdadm --assemble --verbose /dev/md0 /dev/sdb1 /dev/sdc1 /dev/sdd1 mdadm: looking for devices for /dev/md0 mdadm: /dev/sdb1 is identified as a member of /dev/md0, slot 1. mdadm: /dev/sdc1 is identified as a member of /dev/md0, slot 2. mdadm: /dev/sdd1 is identified as a member of /dev/md0, slot 3. mdadm: no uptodate device for slot 0 of /dev/md0 mdadm: added /dev/sdb1 to /dev/md0 as 1 mdadm: added /dev/sdd1 to /dev/md0 as 3 mdadm: added /dev/sdc1 to /dev/md0 as 2 mdadm: /dev/md0 assembled from 2 drives - not enough to start the array. mdadm: /dev/md0 assembled from 2 drives - not enough to start the array (slot 0 is the long-dead drive) the output of mdadm --examine for 2 of the drive (sdc sdd) is similar and looks like this: ... State: Clean Active Devices: 2 Working Devices: 2 Failed Devices: 1 Events: 1923796 ... while the output for sdb looks different: ... State: active Active Devices: 3 Working Devices: 3 Failed Devices: 0 Events: 1923787 ... Note the difference in the Events counter and the state. My guess is that the drive is out of sync with the rest. I tried mdadm --assemble --force --update=summaries to bring the stray Events counter up to date per a recommendation I saw in a forum, but this command segfaults. I tried strace-ing it and it faults right after reading 4K of data from /dev/sdb1. To summarize: I'm not sure what to do next. I've read in forums that I should try to re-create the array but I fear it will completely destroy the data (not sure what creating an array from previously-array-ed disks does). Any help will be appreciated. really! Thanks, -- -- Shachar ___ Linux-il mailing list Linux-il@cs.huji.ac.il http://mailman.cs.huji.ac.il/mailman/listinfo/linux-il
RE: Prive mail-server issues (or: I am not a spammer!)
Hi Itay, Others already mentioned it - spammers have ruined it for us. For a practical solution, and the one I use - order a static IP from your ISP and ask that they update the PTR record for your IP to point to an A record. You could also ask them to perform DNS service for you. Yeah, it's not the cheap-ass solution (should cost you $5-150/month more, depending on the ISP), but it's what you can get away with. -- Shachar -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Itay Duvdevani Sent: Saturday, May 21, 2005 2:46 PM To: IGLU Mailing list Subject: Prive mail-server issues (or: I am not a spammer!) Greetings, I am running a mail server on my private linux box at home, which I use to communicate with the world. I am using a free dynamic DNS service for that purpose (dyndns). As a result, I am unable to send email messages to half the world, since my server gets blacklisted all the time (or the receiving server claims it won't receive mail from dynamic DNS servers). This is really annoying. - This is a private mail server, that has more than one user (so the single mail account I've got from my ISP isn't enough). - It has some filtering installed over it (like spamassain etc), so using a 3rd party mail service (name your free pop/web mail provider) is unacceptable (plus, it doesn't carry my domain name, which is an issue with business mail). Is there a way out of this situation any of you can think of? Thanks, - Itay. == To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
RAID1 root/boot w/Debian 3.0r2
Hi All, I'm trying to set up a machine with the following configuration: /dev/md0 mounted as /boot /dev/md1 mounted as / and more /dev/md* as /usr, /var, etc. once the first two are set up properly. Oh, and I'm using Woody 3.0r2 Most HOWTOs on the net specify that: 1. A floppy should be loaded early during installation (contains raid kernel modules) 2. I should create /dev/md* before installation (using mkraid and having one failed disk configured on purpose), mkfs it, and then install base system on /dev/hd??. 3. After installation I should recompile kernel with RAID built in (not as module). 4. Reboot, reconstruct array, edit lilo.conf with raid-extra-boot and whatnot, run lilo, voila. However, step 4 does not work for me. LILO either doesn't want to install, claiming the /boot/boot.map is not on the boot raid partition, or it installs (if I tweak some other parameters not according to the HOWTO), but then LILO doesn't load (01 lather rinse repeat on the screen). I've also tried installing Woody normally, then build MD devices, cp -a files over, edit fstab, lilo and reboot, but this way lilo claims that this is not a RAID install, and simply refuses to cooperate. I'd appreciate any help/pointers/references. TIA, Shachar = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
[OT] Israeli hosting
Hi All, The company I'm working for is looking for an Israeli (read: located in Gush Dan) hosting solution to enable our customers to send us large files. Storage space required is upwards of 5 GB. Unfortunately, none of the hosting providers we talked to is happy to provide a managed service for anything over 500MB disk space. Recommendations, anyone? -- Bye, Shachar. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
[OT] Sparc CPU Throttling?
Hi fellas, I am suspecting that one of our Sun Blade 1000 (UltraSparc) is throttling its CPUs down during low load periods, much like more recent living-room-heater x86's. Is this technology even available on Sparcs? If so, how do I detect (and change) it? Thanks, Shachar. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: ignoring setgid
Behdad Esfahbod wrote: The only way I can think of is: cp /usr/bin/myprog ~ chmod g-s ~/myprog ~/myprog That, or mounting that file system with nosuid option, but this is probably not what you want :) Shachar. behdad On Fri, 12 Dec 2003, Noam Meltzer wrote: Hi, Assume a situation where I have a program, lets sat: /usr/bin/myprog which the permissions on it are: rwxr-sr-x now, i'm not the owner of the program, but a member of its group. but, i don't want to run this program and getting it change my effective gid. is there a way preventing it from doing so? 10x, Noam = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] . = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: disk problems
Moshe Kaminsky wrote:
Hi,
I have some disk problems: when I try to access certain files, I hear
strange sounds from the hard disk, the computer has a delay, and I get
the following type of messages in /var/log/messages:
Dec 1 23:05:36 localhost kernel: hda: dma_intr: status=0x51 { DriveReady SeekComplete
Error }
Dec 1 23:05:36 localhost kernel: hda: dma_intr: error=0x40 { UncorrectableError },
LBAsect=34897754, sector=1048672
Dec 1 23:05:36 localhost kernel: end_request: I/O error, dev 03:0a (hda), sector
1048672
Dec 1 23:05:36 localhost kernel: EXT3-fs error (device ide0(3,10)):
ext3_get_inode_loc: unable to read inode block - inode=59969, block=131084
Anyone knows what is it, and how can it be fixed?
This is a very strong indication of a bad sector (especially if the
problem persists). If you value your data, you better replace the drive.
Probably even if you don't.
Thanks,
Moshe
Shachar.
=
To unsubscribe, send mail to [EMAIL PROTECTED] with
the word unsubscribe in the message body, e.g., run the command
echo unsubscribe | mail [EMAIL PROTECTED]
Re: [OT?] SPAM
Tzafrir Cohen wrote: On Tue, Nov 25, 2003 at 10:08:51PM +0200, Shachar Tal wrote: Out of the box it does excellent job when it comes to form and format classification (HTML, [EMAIL PROTECTED], etc.), but without training it to *your* ham and spam, it's not too good. How well does SpamAssasin deal with Hebrew mail? E.g: A Hebrew subject is often treated as yelling. Maybe bacause it is quoted-printable? I do not receive much Hebrew mail, and anyway most of it is spam from various companies. Family and friends occasionally send me Hewbew mail, but when they do, SA treat their email address virtually as white-listed, since all mail messages from those addresses is taught to be ham. Shachar. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: [OT?] SPAM
Yuval Aviel wrote: Shachar, I also use spamassassin, but the success rate is much lower. Maybe you can share your .spamassassin/user_prefs with us? This is my user_prefs: required_hits3 That's it. Oh, and I trained SA with a ~16000 message corpus, ~2000 of which were ham, ~14000 were spam. You *did* train it, didn't you? Out of the box it does excellent job when it comes to form and format classification (HTML, [EMAIL PROTECTED], etc.), but without training it to *your* ham and spam, it's not too good. man sa-learn is your friend :) Shachar. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: [OT] Government is switching to OpenOffice
Oleg Goldshmidt wrote: .. still on Windows, but Microsoft is mad nonetheless. http://www.theregister.co.uk/content/4/34154.html Apparently, the feeling is mutual. ;-) No surprise here. The windows OS line generates a small amount of money for MS, compared to the Office product line (I remember reading somewhere that Office comprises more than 50% of MS's profits). Pointers, anyone? Shachar. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
USB ADSL modem - experiences?
Hi all, Does anyone have experience with USB ADSL modem named ALE 070 (purple modem, issued by Bezeq International), under Linux of course? I am offered such a modem and want to make sure it works under Linux. Thanks, Shachar. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: Version control (was: Re: What's wrong with this code?)
Guy Teverovsky wrote: On Tue, 2003-11-18 at 18:58, Oleg Goldshmidt wrote: Tal, Shachar [EMAIL PROTECTED] writes: Easily doesn't mean a sysadmin for a day. Easily means not having to invest considerable man-power into making cvs and diff and branches and IDE integration and nightly building and whatnot work together. YMMV for the definition of considerable. Disclaimer: I have not used ClearCase myself. However, I have an impression that, for one reason or other every company that uses Clear Case also has a full time software configuration *team* whose purpose in life is making ClearCase work for the developers. This does not mean that ClearCase is bad, wrong, or anything. This just means that it probably fits someone's definition of considerable man-power. You can throw a team on ClearCase maintenance, but without first reading the books they will spend all their time poking around in vein. Most of my time spent on ClearCase involves going through the logs to see one more time that it does what it's supposed to do. Oopss... forgot. I do not do it anymore. I have a script that alerts me if something funny is going on. And how much did the time it took you to learn to do that, cost your company? At one company I worked for (about 15 developers) an internal effort was undertaken to write a system for hourly/nightly build of multiple versions of software kept in CVS, at least on two platforms. It took some effort (one person, I don't really remember how much time it took, maybe a week?), but it worked smoothly afterwards. Probably still works, years later - I don't know. Note also that the build system fit the particular development cycle and practices of the outfit - an out-of-the-box solution would not necessarily fit that. It can be setup in ClearCase in 5 minutes. Create a bunch of dynamic views each with it's own brunch and script the hourly/nightly builds inside each view. Couple of one-liners will suffice. What is the cost of weeks work of a decent sysadmin ? I will repeat my last sentence: How much did the time it took you to learn to do that, cost your company? The time it took my sysadmin at previous work to master ClearCase (and teach everybody else the ClearCase Way (tm) to do things) is roughly 2 months her time and ~1 day each person to learn that dreaded UCM. Now, consider this. Just a few days ago a friend, who is a configuration manager for a big and well-known unnamed company, complained informally that ClearCase (which has its own filesystem implemented by Rational as a binary only kernel module) does not co-exist well with that company's corporate standard kernel configuration. And they cannot do anything about it until the vendor (IBM in this case) fixes the problem. I surely hope the vendor will provide a solution in time (until the client's standard kernel changes). Again, this is not as much to criticize ClearCase as to point out that this is something a multibillion dollar company would surely deem considerable. Shachar. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: problems with partrition table
b g wrote: On Tuesday 18 November 2003 17:49, Oron Peled wrote: Sorry for the lame question, but what are magic numbers? Before working on the raw device, I though it would be a good idea to know what I'm doing. Googling for [magic numbers ext3 fdisk] leads to zillions mailing list archives, so if anyone can point me to a resource which has more or less comprehensive explanations, I'll be glad to hear. What I uderstand from Oron's answer is that some number marks the start and the end of a prartition, so by knowing in which blocks they appear, one can reconstruct the partition table. Am I correct? Magic numbers are used to identify the type of filesystem the super block is contained. All EXE files in DOS and derivatives begin with MZ (Mark Z-something was a leader programmer for MS back then), etc. EXT3's super magic is 0xEF53 (see include/linux/ext3_fs.h). I had similar problem (partition table mess) a long time ago (~8 years): snip Ignored numbers which where not in regular intervals (false alarms) what intervals can be considered as regular AFAIK, those regular intervals should be N*2^M, where M is a natural number ~ 13, and N is a small natural number. ext2's interval was very early hardcoded as 8192, I'm not sure how it's done nowadays. If you see 4 super block candidates whose block numbers form an algebraic series, 99.9% that's the right block. You can verify against the rest of the blocks 3. Ran 'fsck -b super_block_number' 4. If all is good (it was in my case), than we can trust the superblock data (start and size of partition) 5. So we can safely use them to fix the partition table (back then I used Norton-Disk-Doctor under DOS to edit the partition table. I think today parted would give you better options -- but didn't check it). Hope it helps, === To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] . Shachar. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
[Fwd: Re: Looking for Quantum HD for desparate data recovery]
Hi Dov, [EMAIL PROTECTED] wrote: Yes, it was bound to happen. My web-server computer at the Weizmann Institute short-circuited and once the smoke cleared out, there was the 10G harddisk with the only copy of lots of data that I am dying to restore... I tried to connect the HD to another computer but it doesn't even spin up. As a desparate last act I have decided to try to find an identical drive and swap the control boards. (It has been done but I have no idea if it will succeed). AFAIK, this will not work. While it *did* work with older disks, newer disks have extra disk space preallocated for remapping bad sectors, moving data around etc., and the mapping table is stored on one of the chips on the circuit board. This means that when swapping the board you will lose the mapping of logical sectors into physical sectors. Reordering ~2^24 sectors can take some time. So, does anyone have any idea of where I can find the following HD? Quantum Fireball CX, 3.5 series 10.2AT P/N CX10A011 Rev 02-B (Found on sticker above cable socket) If anyone has this disk in any of their old boxes I would be more than happy to replace it with a 80G brand new disk of their choice. Thanks in advance, Dov Shachar. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word unsubscribe in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Re: two in one :)
Hi, Zeev's email address is [EMAIL PROTECTED] On Sat, 23 Oct 1999, Uri Bruck wrote: It's documented in "Core PHP Programming" - didn't actually try to throw anything at it yet. Uri On Thu, 21 Oct 1999, Ben-Nes Michael wrote: This is a very good news, I wonder why its not documented anywhere. Any one have the mail of Zeev Suraski so he can add it to the documents ? Ilya Konstantinov wrote: First of all, I have wrote my perfect, or almost perfect, flipping script inPHP, including numbers, english and wrapping. It's quite slow. But that's not the point - Zeev Suraski already did the work for us, though didn't document it anywhere. hebrev(), a built-in function reverses hebrew flawlessly. Just print hebrev($string). Much faster than any PHP-only function doing the same, ofcourse. Actually, you'll find one reference to it, if you go to www.php.net - Credits - click Zeev Suraski. -- Best regards, Ilya Konstantinov a.k.a Toastie [http://toast.demon.co.il] = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] -- -- Canaan Surfing Ltd. Internet Service Providers Ben-Nes Michael - Manager Tel: 972-6-6925757 Fax: 972-6-6925858 http://www.canaan.co.il -- = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED] Shachar Tal - Taub Computer Center, Technion, Israel Institute of Techonlogy finger [EMAIL PROTECTED] for contact information or public key. = To unsubscribe, send mail to [EMAIL PROTECTED] with the word "unsubscribe" in the message body, e.g., run the command echo unsubscribe | mail [EMAIL PROTECTED]
Linux and SGI - a proposed workshop (fwd)
quite interesting... At 18:42 24/05/99 +0300, Moshe Barak wrote: Shalom, I am trying to find out how many people will be interested in a Linux workshop presented by SGI, to take place at the Technion during the week starting 6th June. (exact date and location not established yet). If interested to participate after reading the section below, please follow the instructions at the end of this mail. do NOT send me direct private mail requesting to attend the workshop, thank you! SGI has recently declared full support of Linux and it's contribution of the XFS file system for open systems. The XFS file system enables Linux to support a high level of performance, scalability and stability needed for critical applications, and enables, in case of server failure, a fast recovery without loss of data. Mr. Hank Shiffman, strategic technologist at SGI, specializing in Linux, will present a two hours session during which he will discuss the following subjects: o How will SGI support Linux and how it interconnects with Operating Systems supported today by SGI. o The contribution of SGI's advanced technology to the Open Source Movement Those interested to participate, please do either one of the following: 1. send e-mail to listserv@tx with one line in the mail body: sub tech-linux-l FN SN e.g: sub tech-linux-l Yossi Levy 2. Point your browser to http://tx.technion.ac.il/archives/tech-linux-l.html and click the 'Join/Leave the List' button. As the time frame to have this workshop is very short, please let us know of your intention to participate ASAP. Once we find out if there is enough interest, we will be sending information about the exact time and location. This session is open to all Technion members. Best, Moshe
Re: linux haifa club - we need to find a 'place'
Hi Guy, On Fri, 23 Apr 1999, guy keren wrote: so far there were about a dozen of people that have expressed an interest in participating in such an activity. however, there is yet no place to have the meeting at. someone (hint, hint...) told me he might be able to arrange some place... unless this silence is accidental, i'd like to as if anyone here has the right connections to get us a small 'room' somewhere (a "class"). i'd assume that having some type of blackboard (or white board) is a requirement. having network connection is not a must, thought it could help to set up live demonstrations in the long run. i'd assume that it'l be best to set up such meetings on thurthday's early evening (i.e. 18:00-20:00?) but that's negotiable. I'm not studying in the Technion anymore, but I can arrange a class room in the Technion with a blackboard and a network connection. Shachar Tal - finger [EMAIL PROTECTED] for contact information or public key.
Re: Administrivia (was Re: äöòä äòöä )
At 18:27 30/03/99 +0200, Vadim Vygonets wrote: Quoth Evgeny Stambulchik on Tue, Mar 30, 1999: Vadim Vygonets [EMAIL PROTECTED] wrote: So? Should I setup digests to be sent nightly? It's fine with me. Agreed. If nobody objects. I prefer it to be just before 22:00. Some people dial to their ISPs at 22:00, and do not stay connected till around midnight, having to wake up in the morning, unlike some nightly creatures on this list. :) My opinion is that digests should be sent at 22:00 (if the time can be configured, that is...) Shachar.
Re: qmail relay...
Hi, On Mon, 15 Mar 1999, Ze'ev Maor wrote: I've tried setting my local qmail box to relay mail for local network users. I'm using tcpd, the qmail smpt is called from inetd as instructed by the FAQ, i.e.: smtp stream tcp nowait qmaild /usr/sbin/tcpd /var/qmail/bin/tcp-env /var/qmail/bin/qmail-smtpd (one line ofcourse) And I've also put: tcp-env: 192.168.2.0: setenv = RELAYCLIENT into /etc/hosts.allow but still qmail won't relay... I believe what you want is: tcp-env: 192.168.2.: setenv = RELAYCLIENT so mail from your subnet (192.168.2.x) is relayed, and not the host using 192.168.2.0 as an IP address, if such exists. Shachar Tal - Taub Computer Center, Technion, Israel Institute of Techonlogy finger [EMAIL PROTECTED] for contact information or public key.
