Re: Mixing CPU's
Samat Jain wrote: Lennart Sorensen wrote: On Sat, Jul 23, 2005 at 12:01:27PM +0200, Michelle Konzack wrote: I had a MSI K8D Master3 (380 Euro, Two Opteron 240) which drived me crazy and I have tried to run one of those CPU's on a Singel-Opteron mainboard from Tyan and it does not work. Then I have bought two Tyan Dual-Opteron Boards andused it with only one CPU each, which works very fine. Then I have tried to run a Opteron 142 on a Dual-CPU-Board and it does not work. So you NEED the right CPU to your Mainboard. You can not use a 1xx on a Mainboard which was made for a 2xx. I wonder if that is for technical reasons, or just lack of cpu support in the bioses. The Opteron 1xx series of processors are basically conventional Athlon 64 939-pin processors with an extra pin. I'm not sure if this is CPU related or if its mobo chipset related, but there is a significant difference between the 939's and 940's, as the 939's like standard unregistered memory... 940's will only take reg ecc. Hence the reason they went from the 754 to the 939 since no one wanted to buy the more expensive ram necessary to run it (possibly other reasons too). The Opteron 2xx and 8xx series, however, are different--they both have a different count of what is called coherent HyperTransport links. These links are used for transferring SMP protocol type things (cache coherency checks, cache snooping, etc) between processors. The 2xx series has one of these links (it can connect to one other processor), while the 8xx has three. This is the reason why you can't use a 1xx processor in a 2-CPU system, or a 2xx processor in a 4 or 8-CPU system, but can use an 8xx processor everywhere (the extra coherent links just will not be used). Actually they were reffering to the fact that a 1xx would not work period on a 2xx board. From what I understand the HT pathway is only used when using multiple cpus (ie: cpu to cpu communications). So even if you use one lone 1xx cpu, aparently, its supposed to work on a 2xx board. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mixing CPU's
Latchezar Dimitrov wrote: All that being true however from the arch of duals one should be able to put together dual-core 2-way and a regular, i.e., single-core cpu on a 2-way opteron mobo. Latchezar I guess thats what I'm the most curious about; mixing these dual and single core processors. Do the required Hyper Transport pipelines match up when you use a dual with a single? If so... how do you match the duals with the comparable single cores since it seems they use a different naming scheme (ex: single 244 = = dual 265). It all seems 'theoretically' plausable, I'm just wondering if anyone has actually tried yet. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mixing CPU's
Latchezar Dimitrov wrote: Well, it's true it's theoretical until you try it, however there has been a discussion recently I believe Len Sorensen gave a good description. Also AMD is your friend - go and check it out. You'll see why the cpu are called dual-CORE not dual-cpu Latchezar Awesome, found the thread...very insightful. I was under the impression that there was more than one HT pipeline in the dual cores, but I guess not. Seems like the only issue would be finding how to match them, and as you mentioned before with the AMD site, probably easiest way is to spec out the stats and see what matches. Thanks, Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Mixing CPU's
Has anyone tried mixing the processor types of multi-cpu systems? What was the outcome? Has anyone tried mixing dual-core and single-core cpus together? I'm truely interested in knowing if its necessary to shell out all that money for 2+ chips... Thanks, Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Mixing CPU's
Paolo Alexis Falcone wrote: On 7/23/05, Nathan Dragun [EMAIL PROTECTED] wrote: Has anyone tried mixing the processor types of multi-cpu systems? What was the outcome? Has anyone tried mixing dual-core and single-core cpus together? I'm truely interested in knowing if its necessary to shell out all that money for 2+ chips... Generally it's not a good idea. You've got timing issues to consider, among other things. What kind of timing issues? What other things Just because it isn't recomended dosen't mean it won't work. On the contrary, half the things 'they' tell you won't work will work, they just want to make more money off you. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Has anyone ever....
Ok, I'm curious here. I've got 2x512 sticks of corsair memory for my opteron, and unfortunately they are PC2700 and not PC3200. (Thinking the max FSB was only 333 when I first purchased this system 2+ years ago.) What I'm wondering is; has anyone ever tried overclocking memory on their opterons? Is it fesable? Stability issues? - although thats why I mentioned they were corsair sicks, don't know if that would make a difference or not. In theory I'd like to be able to run them at 3200, but I have no clue if this is considered possible, and I don't have time to pull it down and play with it. Unless I know my efforts are not in vain. Thanks much as always, Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Has anyone ever....
Lennart Sorensen wrote: On Tue, Jun 21, 2005 at 04:51:27PM -0400, Nathan Dragun wrote: Ok, I'm curious here. I've got 2x512 sticks of corsair memory for my opteron, and unfortunately they are PC2700 and not PC3200. (Thinking the max FSB was only 333 when I first purchased this system 2+ years ago.) What I'm wondering is; has anyone ever tried overclocking memory on their opterons? Is it fesable? Stability issues? - although thats why I mentioned they were corsair sicks, don't know if that would make a difference or not. In theory I'd like to be able to run them at 3200, but I have no clue if this is considered possible, and I don't have time to pull it down and play with it. Unless I know my efforts are not in vain. I haven't tried it, but I suspect if you tell it to use them as DDR400, it will either automatically pick slower timins for CAS and such, so it will at best be rather slow DDR400 memory, or you have to make it use slower timings to make it work. I have used PC100 CAS 2 memory as PC133 CAS 3 memory before and that seemed reliable. Len Sorensen Probably very unlikely that it would be stable with the same latencies and timings as the 2700 settings then? I know that some chips far out perform what they boast, where others barely meet the expectations. No clue if Corsair can hold these expections or not. : / Thanks for the reply, Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Has anyone ever....
Jacob Larsen wrote: Corsair is AFAIK a brand that have a pretty good reputation when it comes to overclocking. But experiments are the only way to be sure. You may be lucky that the security margin for you memory modules may allow for them to run PC3200. Thanks, I kinda felt the same way, but I wanted to see if there was anyone else who might think that too. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Has anyone ever (overclocked memory on amd64)...
Hm, unless I'm just used to being able to manually set 200mhz it would appear that I can adjust any and every setting EXCEPT this. I can only limit the frequency...what sense does that make. So right now I'm locked in at 166 it would appear with about 15+ options that I can vary on the memory, which makes me wonder if there is a certan combination of options that I can set to force it into a 200mhz frequency. Any gurus out there have an idea? Thanks, Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Advice sought on moving to AMD64
Lennart Sorensen wrote: Last I checked the 3114 was not very well supported, although that hashopefully improved by now. It is of course software raid only so you are better of pretending it is just a SATA controller and using MD software raid if you want raid at all. I wish I could say how well this driver has worked for me or not, but I strictly use SCSI on this machine anyways, so I'm not sure how good/efficient the driver is. On-board LAN : 2 x Broadcom 10/100/1000 Gigabit Ethernet Are those the tg3 ones? Is there still an issue with firmware in driver without source for those or am I confusing them with something else? Yep this is the TG3, but once again its always worked for me fine. SCSI HDD : 6 x 36GB 15,000 rpm Ultra320 Hmm, expensive stuff. I am not a scsi person anymore. The high prices and lousy performance I got from IBMs 15k rpm scsi drives and raid controller a few years ago while spending a ton of money just makes me not interested anymore. SATA makes much more sense to me. IBM drives have always been unreliable with nicknames such as 'deathstar' for the deskstar series. IBM sold their hard drive division to Hitachi who is now trying to create a new type of storage with deeper platters so that data can be stored vetrically on the disk instead of horizontally end to end. (So from a side view the data looks like ||| instead of --- ) Anyways, its too bad you didn't really see a performance difference for the money you paid, because I'd go SCSI any day. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Advice sought on moving to AMD64
Lennart Sorensen wrote: True, but that may just be a flaw of the built in controller, not of SATA. You can't connect scsi at all to many machines, which doesn't make scsi broken. You can get SATA controllers that run 24 drives if you want, and unlike scsi, they don't even share the connector but instead use one cable per drive. Len Sorensen I hardly see how 24 cables versus 2 is hardly even something you'd have to consider making a choice about. SCSI channels can take one heck of a beating, besides, your point is irrelevant since you can have the same software raid you'd use for SATA over several SCSI channels as well. Lastly, all SATA drives have either IDE or SCSI interfaces anyways before they go to the SATA channel. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Advice sought on moving to AMD64
Jacob Larsen wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Nathan Dragun wrote: Lastly, all SATA drives have either IDE or SCSI interfaces anyways before they go to the SATA channel. Seagate have had native SATA for some time now. Woops, sorry about that bit of mis-information. What kind of performance gains have they been showing over the IDE/SCSI interface SATA drives, any clue? Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Advice sought on moving to AMD64
Rory Campbell-Lange wrote: How hard is it to use Debian AMD64? Hard to use? Wouldn't say hard at all, its like any other OS now. Easy to install in one shot now. Our standard environment includes: Apache 1.3x/Apache 2.x PHP4 Python 2.3/2.4 including all standard modules Perl including all standard modules Postgresql 7.4/8 (I believe I can compile the latter from the debian source packages myself) Exim4 screen netfilter/iptables etc. Everything on your list is available as far as I'm aware, don't know about the postgresql though. You have to realize that its much easier than most to bring the standard x86 packages over since they use the same execution structure. (I think I worded that right, my brain feels fried right now.) One possible configuration we are looking for is as below: Motherboard: Dual AMD Opteron,[X2881G2NR],AMD8131, Up to 3x PCI-X,S-ATA Raid,2x Gigabit LAN Chipset: AMD-8131 Info : 5 x PCI (Total); 1x PCI-X for 1U and 3x PCI-X for 2U; Graphics Slot = None Ports : 2xUSB V2.0 [Rear],PS/2 Kb, Mouse,Serial,Parallel Maximum: RAM 16GB using 8 x 2GB On-board Graphics : Integrated ATI 8MB Rage XL Std HDD Controller : IDE UDMA 100 (Primary Secondary On-board SCSI : None On-board RAID : S-ATA Raid (Raid 0, 1, 10, 4 drives) Silicon Image 3114 On-board LAN : 2 x Broadcom 10/100/1000 Gigabit Ethernet On-board Audio : None CPU: 2 x AMD Opteron 242 1.6GHz (2-way) - 1MB Cache RAM: 2,048 MB Total using 4 x 512MB PC3200 DDR Registered ECC (Use Only In Pairs) Chassis: 2U C215S, 8x H-Swap SCSI Bays, Slim CD and FD bays, 660mm, 2x 64bit PCI, 510W (Black) Rail Kit : Telescopic Rail Kit included with case RAID Controller: LSI MegaRaid 320-1, 64 Bit PCI, Ultra320, 64mb, Single channel, Raid levels 0,1,3,5,10 SCSI HDD : 6 x 36GB 15,000 rpm Ultra320 Ethernet : 1 2 x Broadcom 10/100/1000 Gigabit Ethernet on-board motherboard Actually, I run the S2882UG3NR and have run it almost since day 1 of the x86_64 debian days; works flawlessly. Tyan makes a great motherboard too. Unlike most companies who will put out a motherboard and put out random fixes for flaws, these guys actually go back and meticulously update their supported hardware so that you really do get the most for your money. So in short its a piece of cake. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Successful install, but no boot
J.A. de Vries wrote: On 2005-06-14 @ 16:03:58 (week 24) Clive Menzies wrote: DFS (debian from scratch) is a great Live 'CD' with lots of tools and it boots a grub prompt: http://people.debian.org/~jgoerzen/dfs/html/dfs.html It seems like a very nice tool, but I couldn't get it to work. First I had to edit /etc/dfsbuild/dfs.cfg, because the mirror setting for [repo amd64] was pointing to a non-existent location (may be due to the recent move of the repository). I changed it to mirror = http://debian-amd64.alioth.debian.org/debian-amd64/ which seems to work. Then I commented out the default mirror setting pointing to localhost as it caused a fatal error when trying to rm some files. Now it bails out with: Creating Packages files... Building: unstable dists/unstable/main/binary-amd64 Packages Creating Release files... Building: unstable Release Building: unstable Contents All done, exiting. Running: rm -v /home/hdv/tmp/dfs//target/var/lib/apt/lists/debootstrap.invalid_dists_sid_main_binary-amd64_Packages /home/hdv/tmp/dfs//target/var/lib/apt/lists/debootstrap.invalid_dists_unstable_Release removed `/home/hdv/tmp/dfs//target/var/lib/apt/lists/debootstrap.invalid_dists_sid_main_binary-amd64_Packages' removed `/home/hdv/tmp/dfs//target/var/lib/apt/lists/debootstrap.invalid_dists_unstable_Release' Fatal error: exception Not_found Anyway, I built a Kanotix-64 Live CD and ran grub-install /dev/hda from that. Sadly it didn't solve the problem. The system still sees no boot device there after rebooting. Grx HdV I have this problem as well every time I update grub, it tells me the boot disks are located at (2,0) but they're really located at (0,0). I was able to solve this by manually editing this through the grub boot menu and then saving the changes. Of course I need to do this every time I update grub, but it works. Hope that helps. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ia32-libs broken?
Goswin von Brederlow wrote: Directly to the author of the patch. Read changelog for email. MfG Goswin Thanks Goswin, I ended up doing that a short while back. Although, I've yet to hear back yet. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: ia32-libs broken?
Nathan Dragun wrote: I had issues with pre 1.4 build (non-gcc4) in which the lddlibc4 command couldn't be found. Fortunately in the 1.4 build it worked great! But now suddenly the 1.4.0.0.1.gcc4 build fails to find the command again. I waited a while before submitting this assuming it would draw attention to itself, but it hasn't. If I'm not mistaken the libraries were movied to the ia32-libs-extra package? I grabbed that as well, didn't seem to make a difference in regards to lddlibc4. Still havn't heard word on this, anyone else having an issue with this? If necessary, where/how should I file a bug report? Thanks. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
ia32-libs broken?
I had issues with pre 1.4 build (non-gcc4) in which the lddlibc4 command couldn't be found. Fortunately in the 1.4 build it worked great! But now suddenly the 1.4.0.0.1.gcc4 build fails to find the command again. I waited a while before submitting this assuming it would draw attention to itself, but it hasn't. If I'm not mistaken the libraries were movied to the ia32-libs-extra package? I grabbed that as well, didn't seem to make a difference in regards to lddlibc4. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Lilo problem in gcc4
Kyuu Eturautti wrote: Side question - if this is something troublesome to fix, does anyone have quick tips on switching boot loader from lilo to grub? If I'm not mistaken, its simply a matter of installing the grub package? And then running 'update-grub'. I didn't think there was anything complex about it. Actually, I take that back, I have to go in and maually edit my /boot/grub/menu.lst to target to (hd0,0) since for some reason it wants to link the scsi device to (hd2,0). Highly doubt you'll run into that issue though. Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
Re: Re: SSH package concerns...
but sshd.conf contains the needed flags to limit the authentication methods doing man sshd_config saids something like : UsePAM = yes PasswordAuthentication = no might do the trick PasswordAuthentication is set to no by default, as enabling it causes cleartext password authentication (obviously defeating the point of encrypting it in the first place). And yes UsePAM = yes was set, for clarification. So I'd assume that meant that PAM authentication was final? Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]
SSH package concerns...
While setting up PAM in conjunction with SSH I included the following line to deny access unless found in the following file: authrequiredpam_listfile.so sense=allow onerr=fail item=user file=/etc/sshloginusers Which works, sort of. ...Lets say for examples sake the user bob is trying to get in, but is not listed in this file. Ie: not authorized. If I try to connect via the windows program PuTTY, the first attempt fails, naturally, but if I re-type the password when prompted it will let me in!!! Not good. I tested this several different ways and found that if I try and go from linux box to linux box after about 4 attempts it will let me in. SSH package version: OpenSSH_3.8.1p1 Debian-8.sarge.4 in conjunction with: OpenSSL 0.9.7e 25 Oct 2004 Now I was doing some research into this, figuring I configured something wrong or what not early on when I first noticed this authentication problem existed and noticed that there have been some huge changes from the 3.8.1p1 release back in October 2004 (Ironically if I read that right 4.0 was just released today). Changelog: ftp://ftp.ca.openbsd.org/pub/OpenBSD/OpenSSH/portable/ChangeLog But, why on earth is this package so out of date?? Insight into this would be greatly appreciated. God Bless, Nathan Code is poetry. -- To UNSUBSCRIBE, email to [EMAIL PROTECTED] with a subject of unsubscribe. Trouble? Contact [EMAIL PROTECTED]