Re: different behaviour between 4.x and 5.x (ping response/disk io) [was Re: ]
stheg olloydson wrote: Glad to be of help. Raidframe had been ported to FBSD 5.x, but it was removed because kernel changes broke it, and no one volunteered to fix it. I think gvinum replaced vinum in 5.3 for the same reason. I don't use software raid, so I don't really know. It's a pity that raidframe didn't make it into FreeBSD. It seems to be faster than (g)vinum at least for me. I tried gvinum because i wanted to look on gdbe but it didn't seem to be ripe. It lacked several options from gvinum help and if used not carefully enough it would crash the whole machine. On the other hand vinum seems to be very stable but hasn't GEOM support. But NBSD 2.0 is very nice. A company that I do consulting for has just chosen it as the OS for a new embedded product in their pipeline. I've been toying with the idea of idea of web-enabling my toaster by hooking up temperature sensors and a camera, so web surfers could make toast at my house. Good idea, maybe i should install NetBSD in my hair-drier and have the web crowd decide wether my hairs will be dry or wet. :) I was going to use a 8051 chip and write the code, but maybe I'll use NBSD on an embedded board. (Someday, I may want to add the sprinkler system, pool pump,) A fully controllable house is very nice but it either means a lot of work or a lot of money. It's also quite difficult to upgrade an old house because all the walls and floors would have to be ripped open in order to install cabling. P.S. (to the list in general) Why do all of the questions about FBSD performance, especially 4.x vs 5.x, come from people posting from Windows boxes? Theories? In my case i'm using windows because it's pre-installed on the laptop i use and i never had luck with *bsd/linux on the desktop (and especially laptops). This was meant as a humorous question because of certain Windows users' not well-reasoned or -argued posts on this very topic in the last few days. Trolls are really a plague but i think it's quite calm on this list in regard to some Internet panels. Anyway i hope nobody thought that i'm trolling. :) Different tools for different jobs. Although, in the thread you referenced, you said you were having major SMP problems with NBSD. I hope you get those sorted. Those, I would think, are worse than any network and vinum performance problems. In any event you may wish to check back when 5.4 is released. (No release schedule yet.) There are some more points that made me stick with NetBSD and maybe i'll want to come back to FreeBSD when 5.4 is out. However i think that development for NetBSD will make advances too. What SMP, NetBSD and FreeBSD concerns i have to say that i know much more now than when i wrote to the NetBSD mailing list. I made some assumptions back then which proved to be wrong. Somebody gave me the hint to have a look at sysstat vmstat and i saw that the interrupts grew to an high number on disk i/o. It went a little bit smoother without SMP. I couldn't remember that FreeBSD did so, thus i assumed that it performed better. Checking it again exposed that there was almost the same interrupt behaviour. My apologies for writing emails with unevaluated statements. You're welcome and see you on the 'net. stheg Best Wishes, Jochen Keil ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: different behaviour between 4.x and 5.x (ping response/disk io) [was Re: ]
stheg olloydson wrote: (Sorry about the multiple posts. I somehow sent this without a subject line before.) Never mind. Nobody's perfect. ;) To sum up your problem, you tested FreeBSD 5.3, NetBSD 2.0, FreeBSD 4.11 and an elder version of the Knoppix (Linux 2.4) CD and found that FreeBSD 4.11RC2 had the best ping responses from that group. What you want to know is why FBSD 5.3 doesn't respond as well as 4.11RC2. Is this correct? You got me right. Assuming that it is, the answer is that 5.3 is the first stable release of the 5.x branch. One of the 5.x branch's main purposes is to make FBSD much more scalable in terms SMP support. Doing this requires removing the Giant lock. It had been hoped that the removal process would be finished in time for 5.3. Unfortunately, as often happens in a volunteer project delays occurred for various reasons, that was not the case. The incomplete removal meant that not all subsystems could be optimized properly. One of those subsystems is networking. This is not as bad as it sounds because while 5.3's network performance is not as good as 4.11RC2, it is no worse that of NBSD 2.0 or any Linux distro. Also, the optimization has already begun on networking and 5.4 should be _at least_ as good as 4.x. Also, as you saw yourself, using an SMP kernel in FBSD 5.3 doesn't cause a performance hit in networking but it does in NBSD 2.0. So your choices seem to be use 4.11RC2 (full release due shortly) to get the best network response, 5.3 to get as good performance as NBSD 2.0 but with SMP, or use NBSD 2.0 to get as good perfomance as 5.3 but without SMP. Of course, you can wait until NBSD (your prefered OS) performs as well as FBSD, but that may be a lnng time.:) Just a few day ago i installed NetBSD 2.0 to make a final stroke to my decision as i got this computer on the 16. of november. Main advantage in my opinion is that raidframe performs better than vinum (at least with my setup and with the tests i performed). There are some other topics but mainly subjective and not the matter of this email. I guess i'm going to stick with it, developers and time will do the rest for me. :) But what's most important is that your mail gave me the confidence that my hardware isn't faulty. It's also very nice that you shed some light on the that whole network subject. P.S. (to the list in general) Why do all of the questions about FBSD performance, especially 4.x vs 5.x, come from people posting from Windows boxes? Theories? In my case i'm using windows because it's pre-installed on the laptop i use and i never had luck with *bsd/linux on the desktop (and especially laptops). Did you ever recognize the difference between Firefox/Mozilla Browser for Windows and the versions für FreeBSD? Even only compiled with GTK1 Firefox/Mozilla will load the cpu up to 100% when there are some tabs (10-20) opened. I know that this is a matter of unsupported graphic cards with X.org (that whole proprietary stuff annoys me a lot..) but i don't want to suffer for my beliefs. If there only would be ratpoison or xfce for windows.. Well, enough said. (Maybe i'll make some space free for FreeBSD 5.3 to give a try but i'm not to optimistic that it will suffice my needs) Thanks for your kind assistance and best regards, Jochen Keil ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: different behaviour between 4.x and 5.x (ping response/disk io) [was Re: ]
it was said Just a few day ago i installed NetBSD 2.0 to make a final stroke to my decision as i got this computer on the 16. of november. Main advantage in my opinion is that raidframe performs better than vinum (at least with my setup and with the tests i performed). There are some other topics but mainly subjective and not the matter of this email. I guess i'm going to stick with it, developers and time will do the rest for me. :) But what's most important is that your mail gave me the confidence that my hardware isn't faulty. It's also very nice that you shed some light on the that whole network subject. Glad to be of help. Raidframe had been ported to FBSD 5.x, but it was removed because kernel changes broke it, and no one volunteered to fix it. I think gvinum replaced vinum in 5.3 for the same reason. I don't use software raid, so I don't really know. But NBSD 2.0 is very nice. A company that I do consulting for has just chosen it as the OS for a new embedded product in their pipeline. I've been toying with the idea of idea of web-enabling my toaster by hooking up temperature sensors and a camera, so web surfers could make toast at my house. I was going to use a 8051 chip and write the code, but maybe I'll use NBSD on an embedded board. (Someday, I may want to add the sprinkler system, pool pump,) P.S. (to the list in general) Why do all of the questions about FBSD performance, especially 4.x vs 5.x, come from people posting from Windows boxes? Theories? In my case i'm using windows because it's pre-installed on the laptop i use and i never had luck with *bsd/linux on the desktop (and especially laptops). This was meant as a humorous question because of certain Windows users' not well-reasoned or -argued posts on this very topic in the last few days. Well, enough said. (Maybe i'll make some space free for FreeBSD 5.3 to give a try but i'm not to optimistic that it will suffice my needs) Different tools for different jobs. Although, in the thread you referenced, you said you were having major SMP problems with NBSD. I hope you get those sorted. Those, I would think, are worse than any network and vinum performance problems. In any event you may wish to check back when 5.4 is released. (No release schedule yet.) Thanks for your kind assistance and best regards, Jochen Keil You're welcome and see you on the 'net. stheg __ Do You Yahoo!? Tired of spam? Yahoo! Mail has the best spam protection around http://mail.yahoo.com ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: different behaviour between 4.x and 5.x (ping response/disk io) [was Re: ...
In a message dated 1/16/05 7:43:15 PM Eastern Standard Time, [EMAIL PROTECTED] writes: Now i am really puzzled because i cannot understand why 4.x behaves relatively good compared to 5.x on this specific issue. Is there a good explanation or does one have to investigate this further? Also, as you saw yourself, using an SMP kernel in FBSD 5.3 doesn't cause a performance hit in networking but it does in NBSD 2.0. So your choices seem to be use 4.11RC2 (full release due shortly) to get the best network response, 5.3 to get as good performance as NBSD 2.0 but with SMP, or use NBSD 2.0 to get as good perfomance as 5.3 but without SMP. Of course, you can wait until NBSD (your prefered OS) performs as well as FBSD, but that may be a lnng time.:) Of course you won't be able to run 4.x on the latest hardware, because they've stopped supporting such things. What puzzles me is that they call 5.x the production version, even though they seem to know its not there yet. So (sadly), you can't run the fastest version on the fastest hardware. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: different behaviour between 4.x and 5.x (ping response/disk io) [was Re: ...
In a message dated 1/16/05 7:43:15 PM Eastern Standard Time, [EMAIL PROTECTED] writes: stheg P.S. (to the list in general) Why do all of the questions about FBSD performance, especially 4.x vs 5.x, come from people posting from Windows boxes? Theories? Because performance is a server issue and what you use for a server has little to do with what you use on your desktop. Most people that don't have a religious repulsion with Microsoft use Windows on the desktop. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
different behaviour between 4.x and 5.x (ping response/disk io)
Hi all. I got a dual Piii 800 box here with these specifications: -Supermicro 370DL3 Mainboard -Adaptec aic7892 Ultra160 SCSI adapter (onboard) -2x 256MB Micron ECC RAM (tested with memtest) -2x IBM IC35L036UWDY10-0 (SCSI Drives, only one connected for testing) -LG GSA-4040B (connected via onboard IDE Controller) (a few IDE drives which weren't connected für testing) Full dmesg output is attached to the end of this email. Now to the description of my problem. I tested various Operating systems on this box, FreeBSD 5.3, NetBSD 2.0, FreeBSD 4.11 and an elder version of the Knoppix (Linux 2.4) CD. Every installation except FreeBSD 4.11 showed almost the same behaviour when doing the following test. First i ran iozone on the SCSI drive. While the benchmark was performed i issued a ping to another host in the network. The result drove me insane. Wether with the onboard fxp0 either the em0 i was able to get a stable ping response time which looks like that: i get a bunch (~20-30, depends on disk io) of ping response times around 0.2ms and then there are some really long response times above 1ms (the longest took about 90ms). However FreeBSD 4.x seems to be the only exception to this as it gives me over a complete iozone test only three pings with a response time around 1ms. The other ping times are mostly around 0.2ms with some seldom peaks at 0.5-0.6ms. Due to this behaviour i figured out that this cannot be the fault of the hardware, at least i hope so. A hint from a NetBSD Mailing List pointed me to have a look at the interrupts. On disk IO the interrupt rate on ahc0 is going up to around 300/s. But this is all the same for FreeBSD 4.x and 5.x (i used vmstat -i for that). There also seems to be no difference between using SMP and a single CPU kernel (Information based on FreeBSD 5.3 and NetBSD 2.0 tests as i didn't test this with FreeBSD 4.11 yet). Now i am really puzzled because i cannot understand why 4.x behaves relatively good compared to 5.x on this specific issue. Is there a good explanation or does one have to investigate this further? The NetBSD Mailing List i'm talking about can be found here: http://mail-index.netbsd.org/netbsd-users/2005/01/. (Discussion starts on 05.01.2005: High Load / bad response times). I'm looking forward to your answers and with kind regards, Jochen Keil Dmesg from 4.11: FreeBSD 4.11-RC2 #0: Sun Jan 2 09:47:04 GMT 2005 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC Timecounter i8254 frequency 1193182 Hz CPU: Intel Pentium III (799.62-MHz 686-class CPU) Origin = GenuineIntel Id = 0x683 Stepping = 3 Features=0x383fbffFPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,MMX,FXSR,SSE real memory = 536805376 (524224K bytes) config di pcic0 config di psm0 config di sio1 config di sio0 config di ppc0 config di fdc0 config di bt0 config di ata1 config di aic0 config di aha0 config di adv0 config q avail memory = 516833280 (504720K bytes) Preloaded elf kernel kernel at 0xc0556000. Preloaded userconfig_script /boot/kernel.conf at 0xc055609c. Pentium Pro MTRR support enabled md0: Malloc disk Using $PIR table, 9 entries at 0xc00f5230 npx0: math processor on motherboard npx0: INT 16 interface pcib0: ServerWorks NB6635 3.0LE host to PCI bridge on motherboard pci0: PCI bus on pcib0 pci0: S3 968 graphics accelerator at 3.0 irq 11 fxp0: Intel 82559 Pro/100 Ethernet port 0xd800-0xd83f mem 0xfe80-0xfe8f,0xfe9ef000-0xfe9e irq 9 at device 6.0 on pci0 fxp0: Ethernet address 00:30:48:21:74:63 inphy0: i82555 10/100 media interface on miibus0 inphy0: 10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, auto isab0: ServerWorks IB6566 PCI to ISA bridge at device 15.0 on pci0 isa0: ISA bus on isab0 atapci0: ServerWorks ROSB4 ATA33 controller port 0xffa0-0xffaf at device 15.1 on pci0 ata0: at 0x1f0 irq 14 on atapci0 ata1: at 0x170 irq 15 on atapci0 ohci0: OHCI (generic) USB controller mem 0xfeaff000-0xfeaf irq 10 at device 15.2 on pci0 usb0: OHCI version 1.0, legacy support usb0: OHCI (generic) USB controller on ohci0 usb0: USB revision 1.0 uhub0: (0x1166) OHCI root hub, class 9/0, rev 1.00/1.00, addr 1 uhub0: 4 ports with 4 removable, self powered pcib1: ServerWorks NB6635 3.0LE host to PCI bridge on motherboard pci1: PCI bus on pcib1 ahc0: Adaptec aic7892 Ultra160 SCSI adapter port 0xe800-0xe8ff mem 0xfebff000-0xfebf irq 11 at device 3.0 on pci1 aic7892: Ultra160 Wide Channel A, SCSI Id=7, 32/253 SCBs orm0: Option ROM at iomem 0xc-0xc7fff on isa0 pmtimer0 on isa0 atkbdc0: Keyboard controller (i8042) at port 0x60,0x64 on isa0 atkbd0: AT Keyboard flags 0x1 irq 1 on atkbdc0 kbd0 at atkbd0 vga0: Generic ISA VGA at port 0x3c0-0x3df iomem 0xa-0xb on isa0 sc0: System console at flags 0x100 on isa0 sc0: VGA 16 virtual consoles, flags=0x300 acd0: DVD-R HL-DT-ST DVDRAM GSA-4040B at ata0-master PIO4 Waiting 15 seconds for SCSI devices to settle Mounting root from ufs:/dev/da0s1a da0 at ahc0 bus 0 target 1 lun 0 da0: IBM
Re: different behaviour between 4.x and 5.x (ping response/disk io) [was Re: ]
(Sorry about the multiple posts. I somehow sent this without a subject line before.) it was said: Now i am really puzzled because i cannot understand why 4.x behaves relatively good compared to 5.x on this specific issue. Is there a good explanation or does one have to investigate this further? Hello, First, I do not speak for anyone but myself. Therefore, any information this post contains is based solely on my understanding of that information and is only as accurate as that understanding is correct. To sum up your problem, you tested FreeBSD 5.3, NetBSD 2.0, FreeBSD 4.11 and an elder version of the Knoppix (Linux 2.4) CD and found that FreeBSD 4.11RC2 had the best ping responses from that group. What you want to know is why FBSD 5.3 doesn't respond as well as 4.11RC2. Is this correct? Assuming that it is, the answer is that 5.3 is the first stable release of the 5.x branch. One of the 5.x branch's main purposes is to make FBSD much more scalable in terms SMP support. Doing this requires removing the Giant lock. It had been hoped that the removal process would be finished in time for 5.3. Unfortunately, as often happens in a volunteer project delays occurred for various reasons, that was not the case. The incomplete removal meant that not all subsystems could be optimized properly. One of those subsystems is networking. This is not as bad as it sounds because while 5.3's network performance is not as good as 4.11RC2, it is no worse that of NBSD 2.0 or any Linux distro. Also, the optimization has already begun on networking and 5.4 should be _at least_ as good as 4.x. Also, as you saw yourself, using an SMP kernel in FBSD 5.3 doesn't cause a performance hit in networking but it does in NBSD 2.0. So your choices seem to be use 4.11RC2 (full release due shortly) to get the best network response, 5.3 to get as good performance as NBSD 2.0 but with SMP, or use NBSD 2.0 to get as good perfomance as 5.3 but without SMP. Of course, you can wait until NBSD (your prefered OS) performs as well as FBSD, but that may be a lnng time.:) HTH, stheg P.S. (to the list in general) Why do all of the questions about FBSD performance, especially 4.x vs 5.x, come from people posting from Windows boxes? Theories? __ Do you Yahoo!? Yahoo! Mail - now with 250MB free storage. Learn more. http://info.mail.yahoo.com/mail_250 ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: different behaviour between 4.x and 5.x (ping response/disk io) [was Re: ]
stheg wrote: P.S. (to the list in general) Why do all of the questions about FBSD performance, especially 4.x vs 5.x, come from people posting from Windows boxes? Theories? Cuz if you have freebsd on your desktop, you don't give a damn about its performance. It's just too great. And don't look at me, I'm still waiting for SPDIF support in ALC658 driver to dump my last running Windows for good :-) Very best wishes, Andrew P. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
