Troubleshooting a 3Ware 9550SX under 6.3
I've recently noticed slow performance on a machine that runs amanda and stores backups for our network on a RAID-5. It was running 6.2. I upgraded to 6.3 just today, so it is completely up to date on the 6 branch. Here's hopefully most of the relevant info: 6.3-STABLE FreeBSD 6.3-STABLE #0: Mon Jun 9 12:57:40 MDT 2008 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC amd64 da0 at twa0 bus 0 target 0 lun 0 da0: AMCC 9550SX-12 DISK 3.04 Fixed Direct Access SCSI-3 device da0: 100.000MB/s transfers da0: 2145735MB (4394465280 512 byte sectors: 255H 63S/T 273542C) hw.twa0.driver_version: 3.60.04.003 Unit UnitType Status %Cmpl Stripe Size(GB) Cache AVerify IgnECC -- u0RAID-5OK - 64K 2095.44 OFFOFF OFF Port Status Unit SizeBlocksSerial --- p0 OK u0 698.63 GB 14651491683QD0TAVV p1 NOT-PRESENT - - - - p2 NOT-PRESENT - - - - p3 NOT-PRESENT - - - - p4 OK u0 698.63 GB 14651491683QD0SYE3 p5 NOT-PRESENT - - - - p6 NOT-PRESENT - - - - p7 NOT-PRESENT - - - - p8 OK u0 698.63 GB 14651491683QD0SE36 p9 OK u0 698.63 GB 14651491683QD0THRB p10NOT-PRESENT - - - - p11NOT-PRESENT - - - - Filesystem 1K-blocksUsed Avail Capacity Mounted on /dev/da0s1a4058062 96518 3636900 3%/ devfs1 1 0 100%/dev /dev/da0s2d 1914817666 1647878588 11375366694%/backup /dev/da0s1d4058062 3344 3730074 0%/tmp /dev/da0s1f 1716265387978766 149917650 5%/usr /dev/da0s1e 25385516 124322 23230354 1%/var I'm getting performance like this (no other processes were running at this time): dd if=/dev/zero of=testfile 165760+0 records in 165759+0 records out 84868608 bytes transferred in 25.028232 secs (3390915 bytes/sec) I think I ought to be able to get more than 3.2MB/sec out of this card, right? The machine is about a year old and I am pretty sure it wasn't anywhere near this slow at the beginning, though I don't have any numbers from that time to back up that opinion. What else can I try to get more info or resolve the issue? I've looked online for others complaining of slowness and most of the chatter about twa on freebsd seems to be about the driver having been missing from certain install discs back in the day. No one else seems to be having speed issues. Thanks!! -adam ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: Troubleshooting a 3Ware 9550SX under 6.3
Caching is off; that will kill performance especially for small transactions. If you've got battery backup for this system, consider enabling the onboard cache...? Chuck, that made an immediate and huge difference: # tw_cli set cache c0 u0 on # dd if=/dev/zero of=testfile 5649235+0 records in 5649234+0 records out 2892407808 bytes transferred in 40.800142 secs (70892102 bytes/sec) I had no idea having the cache on would be such a night and day difference. I'm off to order the battery backup card. Thanks!!! -adam ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
memory error with asr-utils built from ports
I've got a 6.0-release machine with the following controller: asr0: Adaptec Caching SCSI RAID mem 0xd000-0xd1ff irq 18 at device 2.1 on pci2 asr0: ADAPTEC 2100S FW Rev. 370F, 1 channel, 256 CCBs, Protocol I2O I've built the asr-utils port and it works great for a few days but then starts giving me this error when I run /usr/local/dpt/raidutil -a d0b0t0d0 Shared Memory Allocation Failed, To Size = 400, From Size = fc00, Errno = 1c Shared Memory Allocation Failed, To Size = 400, From Size = fc00, Errno = 1cEngine connect failed: COMPATILITY number If I reboot the machine it works for a few more days. It's an older controller and oldish software but I would like to be able to keep tans on the array in this machine without having to reboot every few days. Anybody got any ideas? Thanks, Adam ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
SCSI Tape Problems
I have this: sa0 at ahc0 bus 0 target 3 lun 0 sa0: HP C1537A L708 Removable Sequential Access SCSI-2 device Connected to this: ahc0: Adaptec 2940 Ultra SCSI adapter port 0xd000-0xd0ff mem 0xe9511000-0xe9511fff irq 18 at device 10.0 on pci0 aic7880: Ultra Wide Channel A, SCSI Id=7, 16/253 SCBs Running on this: FreeBSD 5.4-RELEASE FreeBSD 5.4-RELEASE #0: Sun May 8 10:21:06 UTC 2005 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC i386 When I try to write to it with amanda I get this in dmesg: ahc0: Recovery Initiated Dump Card State Begins ahc0: Dumping Card State while idle, at SEQADDR 0x7 Card was paused ACCUM = 0x4a, SINDEX = 0x37, DINDEX = 0x24, ARG_2 = 0x3 HCNT = 0x0 SCBPTR = 0x0 SCSISIGI[0x0] ERROR[0x0] SCSIBUSL[0x0] LASTPHASE[0x1]:(P_BUSFREE) SCSISEQ[0x12]:(ENAUTOATNP|ENRSELI) SBLKCTL[0x2]:(SELWIDE) SCSIRATE[0x0] SEQCTL[0x10]:(FASTMODE) SEQ_FLAGS[0xc0]:(NO_CDB_SENT|NOT_IDENTIFIED) SSTAT0[0x5]:(DMADONE|SDONE) SSTAT1[0xa]:(PHASECHG|BUSFREE) SSTAT2[0x0] SSTAT3[0x0] SIMODE0[0x0] SIMODE1[0xa4]:(ENSCSIPERR|ENSCSIRST|ENSELTIMO) SXFRCTL0[0x80]:(DFON) DFCNTRL[0x0] DFSTATUS[0x2d]:(FIFOEMP|DFTHRESH|HDONE|FIFOQWDEMP) STACK: 0x0 0x16a 0x19a 0x3 SCB count = 20 Kernel NEXTQSCB = 6 Card NEXTQSCB = 6 QINFIFO entries: Waiting Queue entries: Disconnected Queue entries: 0:14 QOUTFIFO entries: Sequencer Free SCB List: 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Sequencer SCB Info: 0 SCB_CONTROL[0x44]:(DISCONNECTED|DISCENB) SCB_SCSIID[0x37] SCB_LUN[0x0] SCB_TAG[0xe] 1 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 2 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 3 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 4 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 5 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 6 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 7 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 8 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 9 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 10 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 11 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 12 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 13 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 14 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] 15 SCB_CONTROL[0x0] SCB_SCSIID[0xff]:(TWIN_CHNLB|OID|TWIN_TID) SCB_LUN[0xff]:(SCB_XFERLEN_ODD|LID) SCB_TAG[0xff] Pending list: 14 SCB_CONTROL[0x40]:(DISCENB) SCB_SCSIID[0x37] SCB_LUN[0x0] Kernel Free SCB list: 5 15 16 17 18 19 0 1 2 3 4 7 8 9 13 12 11 10 Untagged Q(3): 14 Dump Card State Ends (sa0:ahc0:0:3:0): SCB 0xe - timed out sg[0] - Addr 0xa30a000 : Length 4096 sg[1] - Addr 0xa8ab000 : Length 4096 sg[2] - Addr 0xa10c000 : Length 4096 sg[3] - Addr 0x9e8d000 : Length 4096 sg[4] - Addr 0x9d0e000 : Length 4096 sg[5] - Addr 0xa3cf000 : Length 4096 sg[6] - Addr 0xa17 : Length 4096 sg[7] - Addr 0xa3f1000 : Length 4096 (sa0:ahc0:0:3:0): Queuing a BDR SCB ahc0: Timedout SCBs already complete. Interrupts may not be functioning. (sa0:ahc0:0:3:0): Bus Device Reset Message Sent (sa0:ahc0:0:3:0): no longer in timeout, status = 24b ahc0: Bus Device Reset on A:3. 1 SCBs aborted (sa0:ahc0:0:3:0): WRITE FILEMARKS. CDB: 10 0 0 0 2 0 (sa0:ahc0:0:3:0): CAM Status: SCSI Status Error (sa0:ahc0:0:3:0): SCSI Status: Check Condition (sa0:ahc0:0:3:0): UNIT ATTENTION asc:29,0 (sa0:ahc0:0:3:0): Power on, reset, or bus device reset occurred (sa0:ahc0:0:3:0): Unretryable error (sa0:ahc0:0:3:0): failed to write terminating filemark(s) I have no clue what that all this is supposed to be indicating to me. This same drive worked fine under BSD/OS 4.something. I installed FreeBSD on the machine just a couple days ago and have been getting this result since then. Sorry for my cluelessness! Hope someone can help. Thanks, Adam ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Recommended SATA Controller on 5.4
I need to find a good SATA controller for use on a 5.4 box. I don't need RAID right away but it might be nice in the future. Anyone had good results with an off the shelf add on card? Thanks, Adam ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
[OT] DNS Administration
Is anybody using web based DNS zone administration for their users? I've found several projects out there, but they either seem to be early betas, or only in french, or much too complex/flexible for the average virtual hosting type customer to understand. Before I roll my own I figured it would be worthwhile to ask. Requirements: Open-source, free Simple for non-technical folks to use Works with Bind Multi-user authenticated (user A can add zones and only edit his zones) Desirable: PHP or Perl-based Thanks, Adam Adam Bayless| vi /etc/mail/aliases Fibernet System Janitor | complaints: /dev/null [EMAIL PROTECTED] | :wq baylessfamily.org/~abayless | newaliases ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: IPsec with racoon
Rick, Thanks for the suggestion, but it is a publicly routable address. It actually appears to be getting all of phase 1 complete and most of phase 2 but just never passes any traffic across the VPN tunnel itself, so I am past the basic connectivity issues. Anyone else have any thoughts? Thanks, Adam At 03:06 PM 10/7/2003, [EMAIL PROTECTED] wrote: Is the external IP address of your VPN device an internet routable IP address? I know that if you are on an ADSL without static IP (like Qwest or MSN adsl) the IP address that is automatically assigned via DHCP by the DSL modem is private IP space, and therefore your VPN will not work. I resorted to getting an Alcatel Speedtouch USB modem and plugging it into a FreeBSD box for my Qwest MSN and set my VPN to go between the 2 FreeBSD boxes. This gave my firewall/gateway a real IP address. Granted, it is dynamic and I have to change my vpn every time my IP address get's re-negotiated, but at least it works. I am trying to figure out a way to dynamicly change the VPN config on both ends when ppp comes up so I don't have to do it manually. Sincerely, Rick Duvall --- Adam Bayless [EMAIL PROTECTED] wrote: I've followed a couple of the tutorials available on the web, including the one in the FreeBSD manual, for setting up an IPsec tunnel between two FreeBSD machines, but I am trying to connect to a netgear VPN device. I'm getting past phase 1 and getting an SA but the traffic will not flow. Without quoting every piece of config, does anybody have any pointers on what might differ between the tutorials on FreeBSD - FreeBSD and talking to a VPN device? Thanks, Adam Adam Bayless| vi /etc/mail/aliases Fibernet System Janitor | complaints: /dev/null [EMAIL PROTECTED] | :wq baylessfamily.org/~abayless | newaliases ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Adam Bayless| vi /etc/mail/aliases Fibernet System Janitor | complaints: /dev/null [EMAIL PROTECTED] | :wq baylessfamily.org/~abayless | newaliases ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: IPsec with racoon
Well, I am the ISP, so I can be sure there are no ports blocked... thanks, Adam At 09:27 AM 10/8/2003, [EMAIL PROTECTED] wrote: You don't have any firewall rules blocking it somewhere in the middle between the two endpoints, do you? Some ISP's will block all traffic except for certain types, but they don't tell you about it. We have a wireless internet provider in town that blocks ports to keep people from using certain types of internet services to save bandwidth. They are an http/email only provider in this sense. VPN will not work across this ISP, regardless of the fact that you have a real IP address with them. I disagree with ISP's doing this if people are paying full price for internet service. However, they charge a very low rate, so people get what they pay for in the end. Sincerely, Rick Duvall --- Adam Bayless [EMAIL PROTECTED] wrote: Rick, Thanks for the suggestion, but it is a publicly routable address. It actually appears to be getting all of phase 1 complete and most of phase 2 but just never passes any traffic across the VPN tunnel itself, so I am past the basic connectivity issues. Anyone else have any thoughts? Thanks, Adam At 03:06 PM 10/7/2003, [EMAIL PROTECTED] wrote: Is the external IP address of your VPN device an internet routable IP address? I know that if you are on an ADSL without static IP (like Qwest or MSN adsl) the IP address that is automatically assigned via DHCP by the DSL modem is private IP space, and therefore your VPN will not work. I resorted to getting an Alcatel Speedtouch USB modem and plugging it into a FreeBSD box for my Qwest MSN and set my VPN to go between the 2 FreeBSD boxes. This gave my firewall/gateway a real IP address. Granted, it is dynamic and I have to change my vpn every time my IP address get's re-negotiated, but at least it works. I am trying to figure out a way to dynamicly change the VPN config on both ends when ppp comes up so I don't have to do it manually. Sincerely, Rick Duvall --- Adam Bayless [EMAIL PROTECTED] wrote: I've followed a couple of the tutorials available on the web, including the one in the FreeBSD manual, for setting up an IPsec tunnel between two FreeBSD machines, but I am trying to connect to a netgear VPN device. I'm getting past phase 1 and getting an SA but the traffic will not flow. Without quoting every piece of config, does anybody have any pointers on what might differ between the tutorials on FreeBSD - FreeBSD and talking to a VPN device? Thanks, Adam Adam Bayless| vi /etc/mail/aliases Fibernet System Janitor | complaints: /dev/null [EMAIL PROTECTED] | :wq baylessfamily.org/~abayless | newaliases ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Adam Bayless| vi /etc/mail/aliases Fibernet System Janitor | complaints: /dev/null [EMAIL PROTECTED] | :wq baylessfamily.org/~abayless | newaliases ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] Adam Bayless| vi /etc/mail/aliases Fibernet System Janitor | complaints: /dev/null [EMAIL PROTECTED] | :wq baylessfamily.org/~abayless | newaliases ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
IPsec with racoon
I've followed a couple of the tutorials available on the web, including the one in the FreeBSD manual, for setting up an IPsec tunnel between two FreeBSD machines, but I am trying to connect to a netgear VPN device. I'm getting past phase 1 and getting an SA but the traffic will not flow. Without quoting every piece of config, does anybody have any pointers on what might differ between the tutorials on FreeBSD - FreeBSD and talking to a VPN device? Thanks, Adam Adam Bayless| vi /etc/mail/aliases Fibernet System Janitor | complaints: /dev/null [EMAIL PROTECTED] | :wq baylessfamily.org/~abayless | newaliases ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
issues with Bind 9.2.1
Bind is dumping cores after a couple of minutes on two of my nameservers running 4.7, while one of them is rock solid. All three boxes are as shown: FreeBSD hostname 4.7-RELEASE FreeBSD 4.7-RELEASE #0: Wed Oct 9 15:08:34 GMT 2002 [EMAIL PROTECTED]:/usr/obj/usr/src/sys/GENERIC i386 Pretty vanilla installs. The config is pretty straightforward too, just a bunch of zones. I've even tried limiting the config to just a hints file and localhost.rev and still get the crash. Here's gdb's take on the core file: GNU gdb 4.18 (FreeBSD) Copyright 1998 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type show copying to see the conditions. There is absolutely no warranty for GDB. Type show warranty for details. This GDB was configured as i386-unknown-freebsd...Deprecated bfd_read called at /usr/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 2627 in elfstab_build_psymtabs Deprecated bfd_read called at /usr/src/gnu/usr.bin/binutils/gdb/../../../../contrib/gdb/gdb/dbxread.c line 933 in fill_symbuf Core was generated by `named'. Program terminated with signal 4, Illegal instruction. Reading symbols from /usr/lib/libc.so.4...done. Reading symbols from /usr/libexec/ld-elf.so.1...done. #0 0x281b3798 in select () from /usr/lib/libc.so.4 (gdb) backtrace #0 0x281b3798 in select () from /usr/lib/libc.so.4 #1 0x8125194 in evloop () at app.c:328 #2 0x81253d8 in isc_app_run () at app.c:526 #3 0x8053e39 in main (argc=1, argv=0xbfbffbf4) at ./main.c:572 #4 0x804a061 in _start () (gdb) quit Both machines show the same trace. Anyone got any ideas for me? Thanks, Adam Adam Bayless| vi /etc/mail/aliases Fibernet System Janitor | complaints: /dev/null [EMAIL PROTECTED] | :wq baylessfamily.org/~abayless | newaliases To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
Re: issues with Bind 9.2.1
Kris, Thanks for the speedy reply. I wondered about hardware issues too, but on two different machines with vastly different hardware and no other indications of problems? The first machine is: CPU: AMD Athlon(tm) Processor (1333.10-MHz 686-class CPU) Origin = AuthenticAMD Id = 0x644 Stepping = 4 Features=0x183fbffFPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,MMX,FXSR AMD Features=0xc044b18,AMIE,DSP,3DNow! real memory = 268369920 (262080K bytes) avail memory = 255954944 (249956K bytes) and the second is: CPU: Pentium II/Pentium II Xeon/Celeron (334.09-MHz 686-class CPU) Origin = GenuineIntel Id = 0x660 Stepping = 0 Features=0x183f9ffFPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,SEP,MTRR,PGE,MCA,CMOV,PAT,PSE36,MMX,FXSR real memory = 268435456 (262144K bytes) avail memory = 256024576 (250024K bytes) I'm following the standard instructions that come with bind for these builds (./configure; make; make install) so am I safe to assume it's gonna build for the right architecture? I've also tried building them from the ports collection with no better results. Any other thoughts? Thanks, Adam At 12:43 PM 11/19/2002 -0800, Kris Kennaway wrote: On Tue, Nov 19, 2002 at 01:28:52PM -0700, Adam Bayless wrote: Core was generated by `named'. Program terminated with signal 4, Illegal instruction. This probably means you either have: a) Compiled the binary for an incorrect architecture or CPU type (e.g. pentium pro running on a pentium) b) something causing memory or filesystem corruption (i.e. bad hardware) that has flipped a bit in the bind file. Kris Adam Bayless| vi /etc/mail/aliases Fibernet System Janitor | complaints: /dev/null [EMAIL PROTECTED] | :wq baylessfamily.org/~abayless | newaliases To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message