[OT] Harvest Moon
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 For astronomy lovers, some interesting tidbits about an event this weekend: Weekend Of The Harvest Moon And The Autumnal Equinox http://www.miamisci.org:8080/ramgen/stargazer/SG0238.rm This is from Jack Horkiemer's website (which really could use a new web designer...) here: http://www.jackstargazer.com/ Enjoy! - -- Derek Martin [EMAIL PROTECTED] - - I prefer mail encrypted with PGP/GPG! GnuPG Key ID: 0x81CFE75D Retrieve my public key at http://pgp.mit.edu Learn more about it at http://www.gnupg.org -BEGIN PGP SIGNATURE- Version: GnuPG v1.0.6 (GNU/Linux) Comment: For info see http://www.gnupg.org iD8DBQE9htO9djdlQoHP510RAvH8AJ9C/lbjVqh6AkXuuu3C4Jl9T88acgCfeOQK BIS09qDTgQ9F9imlCo5M1pE= =42G5 -END PGP SIGNATURE- ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: MELBA meeting next week
In a message dated: Mon, 16 Sep 2002 23:26:50 EDT Derek D. Martin said: At some point hitherto, [EMAIL PROTECTED] hath spake thusly: Once again, I've been too busy to plan a meeting or contact any of those I know who might speak. Don't you pretty much just goof off all day at work? =8^) Well, yeah, but I was on vacation last week, which is the time I would normally have spent getting a speaker lined up :) Anyone have anything they can present on short notice? Or, to pre-empt Paul being too busy next month, does anyone have anything they'd like to talk about at the next meeting, next month? Or subsequently? Or have suggestions for a meeting topic and/or speaker? Actually, next month will be the GNHLUG 3rd Quarterly meeting. I think I have someone lined up for November, but I'll gladly take any volunteers for the future, and in case the Nov. speaker falls through :) -- Seeya, Paul -- It may look like I'm just sitting here doing nothing, but I'm really actively waiting for all my problems to go away. If you're not having fun, you're not doing it right! ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: install questions
One suggestion is that you have a cdrom and a hard drive on each chain. By having both the cdrom and the cd-writer on the same chain, you will run into problems should you try to copy a disk, drive to drive. Also, by putting the hard drives on different chains, you will increase the speed of the copy, vice having them on the same chain. Unfortunately, I have not used System Commander, so I am in no position to comment on its use. Regards, Jeff Kirkland I have an older machine, Pentium 233 MMX, 256MB SDRAM, 512 cache, (2) Western Digital 8.4 GB HDD, 24X CDROM, 2X CDRW. I am presently running it as a Win ME machine, with the (C:\) drive (hda) set as primary master and the (D:\) drive (hdb) set as primary slave. The CDROM is set as the secondary master, with the CDRW set as secondary slave. My intention is to run a tri-boot machine letting Slackware 8.1 and FreeBSD 4.6.2 split the second drive. I also have V-Com's System Commander 7.05 software available. I ran RedHat 5.x a few years ago, successfully sharing a single drive with windoze and using an older version of system commander. 1. Should I change the harddrives so that C:\ is primary master and D:\ is secondary master? 2. Should I use the System Commander software? 3. Should I install Slackware or FreeBSD first? Does it matter? This is a learning/re-learning experience, so I have no problem trying something and going back to scratch...any caveats, criticisms, or advice is appreciated. Thanks, Mike Shields _ Play the Elvis® Scratch Win for your chance to instantly win $10,000 Cash - a 2003 Harley Davidson® Sportster® - 1 of 25,000 CD's - and more! http://r.lycos.com/r/sagel_mail_scratch_tl/http://win.ipromotions.com/lycos_0208 01/index.asp?tc=7087 ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Installing onto RAID
Hi all, I'm playing around with sw RAID under Linux. I have a system which has 4 IDE drives which I want to install onto. I want to mirror all the file systems on all 4 drives, such that if any one drive fails, the system should be able to boot from the next bootable device. However, I seem to be in a catch-22 position. In order to mirror across all the drives, don't I need to install to all the drives? And if so, how do I configure the RAID set *before* I install? Or, is there a way to install to the primary master, yet create a /etc/raidtab file which then gets read and invoked at boot time and will mirror whatever is on /dev/hdaX to /dev/hd[b,c,d]X ? Anyone ever do this before? Thanks, -- Seeya, Paul -- It may look like I'm just sitting here doing nothing, but I'm really actively waiting for all my problems to go away. If you're not having fun, you're not doing it right! ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: Installing onto RAID
Not sure how Debian handles it, but RH allows you to create SW RAID devices during install. You may be able to fake it from Debian by dropping to a shell and creating the RAID partition right after it boots. I'm not sure how it handles the situation of the primary master failing and dropping the MBR. Maybe you can set up LILO/GRUB to install to multiple drives simultaneiously? -Mark On Tue, Sep 17, 2002 at 02:23:59PM -0400, [EMAIL PROTECTED] wrote: Hi all, I'm playing around with sw RAID under Linux. I have a system which has 4 IDE drives which I want to install onto. I want to mirror all the file systems on all 4 drives, such that if any one drive fails, the system should be able to boot from the next bootable device. However, I seem to be in a catch-22 position. In order to mirror across all the drives, don't I need to install to all the drives? And if so, how do I configure the RAID set *before* I install? Or, is there a way to install to the primary master, yet create a /etc/raidtab file which then gets read and invoked at boot time and will mirror whatever is on /dev/hdaX to /dev/hd[b,c,d]X ? Anyone ever do this before? Thanks, -- Seeya, Paul -- It may look like I'm just sitting here doing nothing, but I'm really actively waiting for all my problems to go away. If you're not having fun, you're not doing it right! ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: Installing onto RAID
On Tue, 17 Sep 2002, at 2:23pm, [EMAIL PROTECTED] wrote: I want to mirror all the file systems on all 4 drives ... AFAIK, this is not supported in the Linux md driver. It should be pretty easy to do, but it has not been done. ... such that if any one drive fails, the system should be able to boot from the next bootable device. Keep in mind that, if an IDE device fails, it will often hang the entire IDE bus (or controller (or machine)). Anyway, booting from a mirror set is possible, but you need special support for it. Red Hat, for example, distributes a patched LILO which understands a RAID-1 md device. I have no idea how/if GRUB supports RAID. In order to mirror across all the drives, don't I need to install to all the drives? Yes. And if so, how do I configure the RAID set *before* I install? The same way you do anything before the install -- using the installer. If the installer does not support installing to a software RAID, you may be able to do it manually behind the scenes (i.e., on another virtual console), and then install to the newly created RAID device. But that depends largely on your installer. It would help if we knew what distribution and release you were planning on using. :-) Or, is there a way to install to the primary master, yet create a /etc/raidtab file which then gets read and invoked at boot time and will mirror whatever is on /dev/hdaX to /dev/hd[b,c,d]X ? You can install to a failed mirror, i.e., a mirror set with only one physical device, but you still need to do the RAID setup and install to the RAID device. (Actually, that may not be precisely true. I believe Linux software RAID writes the RAID metadata to the end of the physical device. So, you could, in theory, create a slightly-smaller partition, install to it, then expand the partition, create the failed mirror, and then tell it to re-mirror to the other disk. But that is so kludgey it makes my skin crawl.) -- Ben Scott [EMAIL PROTECTED] | The opinions expressed in this message are those of the author and do not | | necessarily represent the views or policy of any other person, entity or | | organization. All information is provided without warranty of any kind. | ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: Installing onto RAID
[EMAIL PROTECTED] said: Hi all, I'm playing around with sw RAID under Linux. I have a system which has 4 IDE drives which I want to install onto. I want to mirror all the file systems on all 4 drives, such that if any one drive fails, the system should be able to boot from the next bootable device. Hmm, most RAID that I know of mirror to 2 devices. Though I don't see why more couldn't be done. If it can't, why not mirror 2 and have a 3rd and 4th as hot spares? I'd also suggest 1-3 and 2-4 be on seperate controllers to boost reliability. However, I seem to be in a catch-22 position. In order to mirror across all the drives, don't I need to install to all the drives? And if so, how do I configure the RAID set *before* I install? I assume you're trying to RAID root. Surely there's a HOWTO for this? I've done this with Solaris, but not with linux. In any event, the RAID should take care of duping the data to all the devices after you get the partitioning done. -- --- Tom Buskey ___ gnhlug-discuss mailing list [EMAIL PROTECTED] http://mail.gnhlug.org/mailman/listinfo/gnhlug-discuss
Re: Syslog and LOG_LOCALn?
At some point hitherto, Mark Polhamus hath spake thusly: Are there any conventions for the use of syslog facility codes LOG_LOCAL0 thru LOG_LOCAL7? That depends on how you define convention; those facilities are used as generic syslog facilities, allowing for the ability of the system administration staff to very finely control where log output goes. Generally, applications should allow the user (the system administrator) to define what facility they want messages logged to, while picking a reasonable default. This allows for maximum flexibility. Pppd apparently uses LOG_LOCAL2. (Does anyone know if that is compiled in or in some configuration file that I haven't found?). Any other examples of popular software that is using one of the local codes? From the pppd man page: DIAGNOSTICS Messages are sent to the syslog daemon using facility LOG_DAEMON. (This can be overriden by recompiling pppd with the macro LOG_PPP defined as the desired facility.) In order to see the error and debug messages, you will need to edit your /etc/syslog.conf file to direct the mes sages to the desired output device or file. I was suprised to learn there were only 24 codes, I just thought the facility identifier would be a string. You're not counting them correctly -- the facility and the level are chosen independently. There are 16 facilities, and 8 levels, thus there are 16 x 8 or 128 different combinations. It is also possible that the number of facilities may be different on some systems. Not all variants of Unix, IIRC, support both the AUTH and AUTHPRIV facilities. Some variants may define others... I'm writing a backup utility. I think I would be best to use syslog, except maybe for larger output which it could write to a file in /var/log/. Does that sound right? I'll make the facility code configurable. There's very little point to logging seperately to a file, particularly if you're just going to put it in /var/log anyway... just let syslog do the work for you. I.E. log the larger output to a different level and/or facility, and let the sysadmin configure where to send that info... You might want to provide the ability to have the admin configure the utility to NOT use syslog, but doing so (that is, the configuring -- not the providing) is generally regarded as bad practice, as it defeats the whole purpose of having a central syslog server. IMNSHO, best not to encourage bad practices... For user processes, this doesn't apply, as users often don't have access to read the system logs. But for system processes (like back-up utilities), generally syslog is the way to go. Aside from that, yes; this does sound right. -- Derek Martin [EMAIL PROTECTED] - I prefer mail encrypted with PGP/GPG! GnuPG Key ID: 0x81CFE75D Retrieve my public key at http://pgp.mit.edu Learn more about it at http://www.gnupg.org msg00685/pgp0.pgp Description: PGP signature
