RE: Dropped Packets
Hi all, I have done some more investigation and stumbled into something odd. If I use ping currently I am getting about 3% packet loss (it gets worse) but with hping (to the exact same ip) I get no loss? As I understand hping and ping use different methods of sending packets, could this be an indication to where my problem might be (buffering or something) ? Ping -- 290 packets transmitted, 281 packets received, 3% packet loss Hping -- 289 packets tramitted, 289 packets received, 0% packet loss I'm using icmp mode in hping, and sending to the same ip... Thanks again for the help Dave -Original Message- From: Dave Raven [mailto:[EMAIL PROTECTED] Sent: Sunday, February 24, 2008 6:25 PM To: 'Simon Chang' Cc: 'freebsd-questions@freebsd.org' Subject: RE: Dropped Packets Hi, Its interesting you'd say that I'd virtually ruled it out given that it takes over a day to start? What I do see sometimes is a message warning me that its limiting open port RST responses - the unit is under a reasonable amount of load though (but not overloaded). The strange thing is that its open port (not closed). Even if I up the limit though (the messages do stop) it doesn't stop the dropping... I've rebooted it recently, but here is the dmesg -- CPU: Intel(R) Xeon(R) CPU5150 @ 2.66GHz (2666.68-MHz 686-class CPU) Origin = GenuineIntel Id = 0x6fb Stepping = 11 Features=0xbfebfbffFPU,VME,DE,PSE,TSC,MSR,PAE,MCE,CX8,APIC,SEP,MTRR,PGE,MCA ,CMOV,PAT,PSE36,CLFLUSH,DTS,ACPI,MMX,FXSR,SSE,SSE2,SS,HTT,TM,PBE Hyperthreading: 2 logical CPUs real memory = 3489005568 (3407232K bytes) avail memory = 3395510272 (3315928K bytes) Preloaded elf kernel kernel at 0xc0378000. Preloaded elf module splash_bmp.ko at 0xc037809c. Preloaded splash_image_data /boot/optec.bmp at 0xc0378140. Preloaded elf module if_em.ko at 0xc037818c. Preloaded elf module if_silbpi.ko at 0xc037822c. Preloaded elf module hptmv.ko at 0xc03782d0. Pentium Pro MTRR support enabled md0: Malloc disk Using $PIR table, 31 entries at 0xc00fddd0 npx0: math processor on motherboard npx0: INT 16 interface pcib0: Host to PCI bridge on motherboard pci0: PCI bus on pcib0 pcib8: PCI to PCI bridge (vendor=8086 device=25f7) irq 0 at device 2.0 on pci0 pci1: PCI bus on pcib8 pcib9: PCI to PCI bridge (vendor=8086 device=3500) irq 11 at device 0.0 on pci1 pci2: PCI bus on pcib9 pcib10: PCI to PCI bridge (vendor=8086 device=3510) irq 11 at device 0.0 on pci2 pci3: PCI bus on pcib10 pcib11: PCI to PCI bridge (vendor=8086 device=0329) at device 0.0 on pci3 pci4: PCI bus on pcib11 pcib12: PCI to PCI bridge (vendor=8086 device=032a) at device 0.2 on pci3 pci5: PCI bus on pcib12 hptmv0: RocketRAID 182x SATA Controller mem 0xd800-0xd807 irq 11 at device 1.0 on pci5 RocketRAID 182x SATA Controller driver Version 1.12 RR182x [0,0]: channel started successfully RR182x [0,1]: channel started successfully RR182x [0,2]: channel started successfully RR182x [0,3]: channel started successfully RR182x [0,4]: channel started successfully RR182x [0,5]: channel started successfully RR182x [0,6]: channel started successfully RR182x [0,7]: channel started successfully RR182x: RAID5 write-back enabled pcib13: PCI to PCI bridge (vendor=8086 device=3518) irq 5 at device 2.0 on pci2 pci6: PCI bus on pcib13 pci6: unknown card (vendor=0x8086, dev=0x1096) at 0.0 irq 5 pci6: unknown card (vendor=0x8086, dev=0x1096) at 0.1 irq 10 pcib14: PCI to PCI bridge (vendor=8086 device=350c) at device 0.3 on pci1 pci7: PCI bus on pcib14 silbpi0: PXG2BPIG port 0x3000-0x303f mem 0xd820-0xd821 irq 11 at device 1.0 on pci7 silbpi0: Speed:N/A Duplex:N/A silbpi1: PXG2BPIG port 0x3040-0x307f mem 0xd822-0xd823 irq 7 at device 1.1 on pci7 silbpi1: Speed:N/A Duplex:N/A pcib15: PCI to PCI bridge (vendor=8086 device=25f8) irq 0 at device 4.0 on pci0 pci8: PCI bus on pcib15 pcib16: PCI to PCI bridge (vendor=8086 device=25f9) irq 0 at device 6.0 on pci0 pci9: PCI bus on pcib16 pcib17: PCI to PCI bridge (vendor=8086 device=2690) irq 7 at device 28.0 on pci0 pci10: PCI bus on pcib17 pcib18: Intel 82801BA/BAM (ICH2) Hub to PCI bridge at device 30.0 on pci0 pci11: PCI bus on pcib18 pci11: ATI model 515e graphics accelerator at 1.0 irq 5 isab0: PCI to ISA bridge (vendor=8086 device=2670) at device 31.0 on pci0 isa0: ISA bus on isab0 atapci0: Generic PCI ATA controller port 0x1800-0x180f,0-0x3,0-0x7,0-0x3,0-0x7 at device 31.1 on pci0 ata0: at 0x1f0 irq 14 on atapci0 ata1: at 0x170 irq 15 on atapci0 pci0: unknown card (vendor=0x8086, dev=0x269b) at 31.3 irq 10 pcib1: Host to PCI bridge on motherboard pci12: PCI bus on pcib1 pcib2: Host to PCI bridge on motherboard pci13: PCI bus on pcib2 pcib3: Host to PCI bridge on motherboard pci14: PCI bus on pcib3 pcib4: Host to PCI bridge on motherboard pci15: PCI bus on pcib4 pcib5: Host to PCI bridge on motherboard pci16: PCI bus on pcib5 pcib6: Host to PCI bridge on motherboard pci17: PCI bus on pcib6 pcib7: Host to PCI bridge
Dropped Packets
Hi all, I have a FreeBSD 4.11 (I know its outdated but I don't have an option) box which is behaving very weirdly. After about a day and a half it starts to drop packets - about 6-10% of pings at least, and tcp traffic becomes unreliable etc. I have swapped out the network cards, the cables etc and I don't have any mbuf problems and all seems fine. For the first day the box works fine - pinging, doing tcp traffic etc; and then suddenly it just stops. Rebooting brings it all back to normal. Any ideas what could be causing this, and how I could go about diagnosing it? As far as I know there are no known bugs for this. I'm using the em driver, but not with SMP. There are no errors on the interface, and the em debug_info and stats sysctl's don't show any problems - there is no indication on the box itself that its dropping packets.. My thinking is that it must be box specific as a reboot solves the problem? Thanks in advance Dave ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Dropped Packets
ad0: 76319MB ST380215A [155061/16/63] at ata0-master BIOSDMA da0 at hptmv0 bus 0 target 0 lun 0 da0: ST380815 AS 3.AA Fixed Direct Access SCSI-0 device da0: 76319MB (156301477 512 byte sectors: 255H 63S/T 9729C) da1 at hptmv0 bus 0 target 1 lun 0 da1: ST380815 AS 3.AA Fixed Direct Access SCSI-0 device da1: 76319MB (156301477 512 byte sectors: 255H 63S/T 9729C) da2 at hptmv0 bus 0 target 2 lun 0 da2: ST380815 AS 3.AA Fixed Direct Access SCSI-0 device da2: 76319MB (156301477 512 byte sectors: 255H 63S/T 9729C) da3 at hptmv0 bus 0 target 3 lun 0 da3: ST380815 AS 3.AA Fixed Direct Access SCSI-0 device da3: 76319MB (156301477 512 byte sectors: 255H 63S/T 9729C) da4 at hptmv0 bus 0 target 4 lun 0 da4: ST380815 AS 3.AA Fixed Direct Access SCSI-0 device da4: 76319MB (156301477 512 byte sectors: 255H 63S/T 9729C) da5 at hptmv0 bus 0 target 5 lun 0 da5: ST380815 AS 3.AA Fixed Direct Access SCSI-0 device da5: 76319MB (156301477 512 byte sectors: 255H 63S/T 9729C) da6 at hptmv0 bus 0 target 6 lun 0 da6: ST380815 AS 3.AA Fixed Direct Access SCSI-0 device da6: 76319MB (156301477 512 byte sectors: 255H 63S/T 9729C) da7 at hptmv0 bus 0 target 7 lun 0 da7: ST380815 AS 3.AA Fixed Direct Access SCSI-0 device da7: 76319MB (156301477 512 byte sectors: 255H 63S/T 9729C) Mounting root from ufs:/dev/ad0s1a Thanks for the help Dave -Original Message- From: Simon Chang [mailto:[EMAIL PROTECTED] Sent: Sunday, February 24, 2008 5:57 PM To: Dave Raven Cc: freebsd-questions@freebsd.org Subject: Re: Dropped Packets Hardware issues come to mind. Do you have a dmesg? SC On Sun, Feb 24, 2008 at 9:27 AM, Dave Raven [EMAIL PROTECTED] wrote: Hi all, I have a FreeBSD 4.11 (I know its outdated but I don't have an option) box which is behaving very weirdly. After about a day and a half it starts to drop packets - about 6-10% of pings at least, and tcp traffic becomes unreliable etc. I have swapped out the network cards, the cables etc and I don't have any mbuf problems and all seems fine. For the first day the box works fine - pinging, doing tcp traffic etc; and then suddenly it just stops. Rebooting brings it all back to normal. Any ideas what could be causing this, and how I could go about diagnosing it? As far as I know there are no known bugs for this. I'm using the em driver, but not with SMP. There are no errors on the interface, and the em debug_info and stats sysctl's don't show any problems - there is no indication on the box itself that its dropping packets.. My thinking is that it must be box specific as a reboot solves the problem? Thanks in advance Dave ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
2TB (and above) Disk
Hi all, Firstly I am using FreeBSD 4, I know it's not supported but hopefully someone has some experience with large disks on 4. I have a box with 6x 500gig drives, configured in raid 5 which comes out around 2.3 terabytes of space. With that, the system picks up the drive (da0) as 0MB in size. I haven't been able to solve that, but took one drive out (moving it to 1.9gig) and it works fine... However, when I try to use /stand/sysinstall I get tons of drive geometry errors, and can't create more than a 1TB slice. If I use A to use whole disk, it goes through with a negative number but then the labeling part of the process gives a paritions are larger than actual chunk error. I then ran dd if=/dev/zero of=/dev/da0 bs=1m count=1 and disklabel -wr da0 auto. This seems to setup a working paritition table, but any attempts on the drive after that to do a newfs result in: newfs preposterous size and a negative number... Any ideas or is it even possible? Thanks in advance Dave ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Low Priority Apps
Hi all, I have a FreeBSD 4.x box running an app that uses 30-50% of the processor. Periodically (through cron) we run a perl app that processes log files and inserts the entries to sql. This perl + mysql combination is causing problems for our app and I'd like to run them at a low priority. What nice value should I give them - should I lower the value of just the perl, or both, or make the high priority app's value higher? Thanks Dave ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Cisco/BSD auto-negotiation
Hi all, I'm currently looking into a problem with a Cisco 3640 router and a FreeBSD 4.9 unit, connected via a crossover cable, that are not negotiating correctly. If you force the media setting to full or half duplex it has constant collisions on the interface, and if you let both autonegotiate the cisco keeps resetting (every ~20 seconds) its network card. Are there known issues with this, or any known fixes? Thanks Dave Ps. Please copy me on replies as I am off the list. ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Bridging a Cisco Trunk
Hi Peter and list, I am unfortunately using 4.x - but it should work fine as far as my understanding is. I'm not sure why it isn't working, but if I bridge em0 and em1 (my two interfaces) the cisco switches can ping each other. The problem is that anything inside the vlan being trunked doesn't go through Now I have tried your setup below and created the vlans and bridged them, which didn't work, but I don't understand the need for the vlans to be created anyway. Surely I should be able to just transparently bridge any traffic that comes from the one side through to the other, and it should work fine? I don't need to communicate on the vlan, just bridge anything that comes through... If I do have to add vlan0,vlan1 to a bridge that's also fine - but at the moment its not working. Thanks for the help Dave -Original Message- From: Peter Wood [mailto:[EMAIL PROTECTED] Sent: 11 January 2006 08:29 PM To: Dave Raven Cc: freebsd-questions@freebsd.org Subject: Re: Bridging a Cisco Trunk Dave, I have two cisco switches, configured to put ports 2-6 on each of them into vlan 100. Then I have port 1 on both set to trunk between the two switches. If I have a device on port 2 on switch1 it can ping a device on port 2 on switch2. I do this quite often, and it works very well on 6.0 for me. You haven't mentioned what version your using, but I will assume you have if_bridge. If you don't and you're gonna use this machine alot for bridging, I'd recommend moving to 6.0. So presumably, you have two interfaces, plugged into the trunk port on each cisco. For arguements sake, we'll say you have an fxp0 and fxp1. So first step is you need to make sure these two interfaces are up, very important, if they arn't, then it wont work. It's easy to forget if you arn't assigning IP's to them. Remove polling if you don't have it compiled into the kernel, but again if you're gonna be bridging packets alot, get it compiled in. It helps alot. ifconfig_fxp0=up polling ifconfig_fxp1=up polling Now create the vlans (and the bridge for later on). cloned_interfaces=vlan0 vlan1 bridge0 ifconfig_vlan0=vlan 100 vlandev fxp0 up ifconfig_vlan1=vlan 100 vlandev fxp1 up In the above please note the ups, if they arn't up then it wont bridge. Now setup the bridge, again noticing the up. ifconfig_bridge0=addm vlan0 addm vlan1 up It should now be working, watch the kernel console and the cisco's logs to see if there are any mismatches or bridging loops. It also seems that you have to put the up at the end of these commands, it took an hour of debugging last night after I had put the up at the start of the ifconfig_vlan lines. Give it a go, send a reply to both me and the list if you are still stuck, Pete. -- Peter Wood :: [EMAIL PROTECTED] ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Bridging a Cisco Trunk
Hi all, I have two cisco switches, configured to put ports 2-6 on each of them into vlan 100. Then I have port 1 on both set to trunk between the two switches. If I have a device on port 2 on switch1 it can ping a device on port 2 on switch2. If I break the link between the two switches, and try to bridge that trunk with a freebsd box, I can't get it right. Does anyone have any specific advice? As I understand it I should just be able to bridge my two interfaces, I have created vlan100 interfaces bound to each though and bridged them as well just to be sure - neither option works... Any advice? Thanks in advance Dave P.s. please copy me as I'm not on the list ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Bridging VLAN's
Hi all, I've done some research on bridging vlans and can't get it right with FreeBSD bridge. What I want to do is bridge an undefined number of vlans through a BSD machine. For example. Vlan 10 from em0 out em1. Now I can't create each vlan and bridge those, because you can't have a vlan10 bound to em0 and to em1, if you create different ones and bridge them the packet comes in on the right vlan but leaves tagged for the wrong one. I read a cisco book that suggests you can bridge normally (just em0,em1) if you set the mtu to 1496, which didn't work. I also googled someone saying 1504 - also not working. Does anyone have any advice? Thanks Dave ___ freebsd-questions@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Datasize change
Hi all, I'm having a problem where I need to allow a proccess to use up to 1.5gig of memory, however its going to be a problem to recompile my kernel. I know that these options would solve the problem: options MAXDSIZ = (1536 * 1024 * 1024) options DFLDSIZ = (1536 * 1024 * 1024) However, I'm wondering if there is a way to do it with login.conf (limits) - here you can see the limits when datasize is unspecified (using kernel default) - however could I increase it through that mechanism? Or is there another way I could do this... Resource limits (current): cputime infinity secs filesize infinity kb datasize 524288 kb stacksize 65536 kb coredumpsize infinity kb memoryuseinfinity kb memorylocked infinity kb maxprocesses 7390 openfiles 32768 sbsize infinity bytes vmemoryuse infinity kb Thanks in advance Dave ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Urgent 4.9 networking problems
Hi all, I really need some urgent help with this I'm completely confused. I have a FreeBSD 4.9 machine running ipfilter ipnat vrrp and a few other services, today is the first time I tried to access through the specific method but now every interface and every local address I try has the same problem. I can ping anything - but any other kind of traffic waits for about 2 minutes before transmitting - this is true with tcp and udp. I'm trying to access machines on the same network - and if I ping -R you can see the same effect - pasted below. I've also included the interface that I'm trying to do this on although it seems to be happening on all my other interfaces.. I try to telnet to a cisco router that's on a switch I'm plugged in and I see the same behaviour - it just waits then suddenly responds very quickly. My IpFilter rules don't log anything until it responds at which time they pass it - and tethereal + tcpdump also see if perfectly AFTER the long delay. It appears that its sitting on the kernel for 2 minutes??? It just does NOTHING then all of a sudden responds. The only thing I can find that works is icmp - and perfectly. I'm sorry for the urgency but its very high priority Thanks in advance Dave # ifconfig fxp1 fxp1: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST mtu 1500 inet x.y.186.3 netmask 0xff00 broadcast x.y.186.255 inet x.y.186.1 netmask 0x broadcast x.y.186.1 inet x.y.186.15 netmask 0x broadcast x.y.186.15 inet x.y.186.14 netmask 0x broadcast x.y.186.14 inet x.y.186.142 netmask 0x broadcast x.y.186.142 inet x.y.186.33 netmask 0x broadcast x.y.186.33 inet x.y.186.124 netmask 0x broadcast x.y.186.124 inet x.y.186.250 netmask 0x broadcast x.y.186.250 inet x.y.186.122 netmask 0x broadcast x.y.186.122 inet x.y.186.25 netmask 0x broadcast x.y.186.25 inet x.y.186.127 netmask 0x broadcast x.y.186.127 # date ; ping -R -c1 x.y.186.253 ; date Thu Jun 24 22:43:13 SAST 2004 PING x.y.186.253 (152.110.186.253): 56 data bytes 64 bytes from x.y.186.253: icmp_seq=0 ttl=255 time=0.414 ms RR: x.y.186.253 x.y.186.253 x.y.186.3 --- x.y.186.253 ping statistics --- 1 packets transmitted, 1 packets received, 0% packet loss round-trip min/avg/max/stddev = 0.414/0.414/0.414/0.000 ms Thu Jun 24 22:46:58 SAST 2004 ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Urgent 4.9 networking problems
Okay thank you I've been so confused with all this that it didn't even occur to me - its now responding as expected - but I still have my original TCP problem.. It takes EXTREMELY long to send the first SYN, once its done that the entire session is perfect... Anyone at all? Any suggestions on further tests? Thanks again Dave -Original Message- From: Gordon Freeman [mailto:[EMAIL PROTECTED] Sent: 24 June 2004 11:09 PM To: Dave Raven Subject: Re: Urgent 4.9 networking problems try ping -nR -c1 x.y.186.254 If you don't get the same lag then it is your DNS lookup that is causing the problem. On Thu, 24 Jun 2004 22:54:10 +0200, Dave Raven [EMAIL PROTECTED] wrote: Hi all, I really need some urgent help with this I'm completely confused. I have a FreeBSD 4.9 machine running ipfilter ipnat vrrp and a few other services, today is the first time I tried to access through the specific method but now every interface and every local address I try has the same problem. I can ping anything - but any other kind of traffic waits for about 2 minutes before transmitting - this is true with tcp and udp. I'm trying to access machines on the same network - and if I ping -R you can see the same effect - pasted below. I've also included the interface that I'm trying to do this on although it seems to be happening on all my other interfaces.. I try to telnet to a cisco router that's on a switch I'm plugged in and I see the same behaviour - it just waits then suddenly responds very quickly. My IpFilter rules don't log anything until it responds at which time they pass it - and tethereal + tcpdump also see if perfectly AFTER the long delay. It appears that its sitting on the kernel for 2 minutes??? It just does NOTHING then all of a sudden responds. The only thing I can find that works is icmp - and perfectly. I'm sorry for the urgency but its very high priority Thanks in advance Dave # ifconfig fxp1 fxp1: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST mtu 1500 inet x.y.186.3 netmask 0xff00 broadcast x.y.186.255 inet x.y.186.1 netmask 0x broadcast x.y.186.1 inet x.y.186.15 netmask 0x broadcast x.y.186.15 inet x.y.186.14 netmask 0x broadcast x.y.186.14 inet x.y.186.142 netmask 0x broadcast x.y.186.142 inet x.y.186.33 netmask 0x broadcast x.y.186.33 inet x.y.186.124 netmask 0x broadcast x.y.186.124 inet x.y.186.250 netmask 0x broadcast x.y.186.250 inet x.y.186.122 netmask 0x broadcast x.y.186.122 inet x.y.186.25 netmask 0x broadcast x.y.186.25 inet x.y.186.127 netmask 0x broadcast x.y.186.127 # date ; ping -R -c1 x.y.186.253 ; date Thu Jun 24 22:43:13 SAST 2004 PING x.y.186.253 (152.110.186.253): 56 data bytes 64 bytes from x.y.186.253: icmp_seq=0 ttl=255 time=0.414 ms RR: x.y.186.253 x.y.186.253 x.y.186.3 --- x.y.186.253 ping statistics --- 1 packets transmitted, 1 packets received, 0% packet loss round-trip min/avg/max/stddev = 0.414/0.414/0.414/0.000 ms Thu Jun 24 22:46:58 SAST 2004 ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Urgent 4.9 networking problems
I have made further progress - thanks for all your steady replies. I know it might look like I haven't looked into it enough but this is just part of my bigger problem - here we go.. By adding my routers ip and my local machines ip to hosts, I've fixed the telnet to the router and the ping -R - but why is telnet timing out ?? I have NO DNS at all - there is nothing in resolv.conf yet it still makes requests to local host. I have to disable dns. I have no idea why it would sit for 2 minutes trying to resolve the ip for my telnet though??? Is this a problem? How do I stop dns altogether... The machine is acting as a firewall with NAT'ing and routing. The real problem that's gotten me down to here is with IPNat though - it says its map'd the address but in actual fact freebsd forwards it. Could this all be a red herring as a dns problem? Thanks Dave -Original Message- From: JJB [mailto:[EMAIL PROTECTED] Sent: 24 June 2004 11:23 PM To: Dave Raven; [EMAIL PROTECTED] Subject: RE: Urgent 4.9 networking problems Your symptoms are typical of DNS time outs. Ping ip address does no DNS lookups. Ping freebsd.org will not work either. With out a lot more detail about your network environment, the best I can say is look at how your network resolves DNS lookups. Some times a ISP will change the ip address of their DNS or DHCP servers and if you have their ip address hard coded in your firewall rules your network will just stop talking to the public internet. Start your research there. -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] Behalf Of Dave Raven Sent: Thursday, June 24, 2004 4:54 PM To: [EMAIL PROTECTED] Subject: Urgent 4.9 networking problems Hi all, I really need some urgent help with this I'm completely confused. I have a FreeBSD 4.9 machine running ipfilter ipnat vrrp and a few other services, today is the first time I tried to access through the specific method but now every interface and every local address I try has the same problem. I can ping anything - but any other kind of traffic waits for about 2 minutes before transmitting - this is true with tcp and udp. I'm trying to access machines on the same network - and if I ping -R you can see the same effect - pasted below. I've also included the interface that I'm trying to do this on although it seems to be happening on all my other interfaces.. I try to telnet to a cisco router that's on a switch I'm plugged in and I see the same behaviour - it just waits then suddenly responds very quickly. My IpFilter rules don't log anything until it responds at which time they pass it - and tethereal + tcpdump also see if perfectly AFTER the long delay. It appears that its sitting on the kernel for 2 minutes??? It just does NOTHING then all of a sudden responds. The only thing I can find that works is icmp - and perfectly. I'm sorry for the urgency but its very high priority Thanks in advance Dave # ifconfig fxp1 fxp1: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST mtu 1500 inet x.y.186.3 netmask 0xff00 broadcast x.y.186.255 inet x.y.186.1 netmask 0x broadcast x.y.186.1 inet x.y.186.15 netmask 0x broadcast x.y.186.15 inet x.y.186.14 netmask 0x broadcast x.y.186.14 inet x.y.186.142 netmask 0x broadcast x.y.186.142 inet x.y.186.33 netmask 0x broadcast x.y.186.33 inet x.y.186.124 netmask 0x broadcast x.y.186.124 inet x.y.186.250 netmask 0x broadcast x.y.186.250 inet x.y.186.122 netmask 0x broadcast x.y.186.122 inet x.y.186.25 netmask 0x broadcast x.y.186.25 inet x.y.186.127 netmask 0x broadcast x.y.186.127 # date ; ping -R -c1 x.y.186.253 ; date Thu Jun 24 22:43:13 SAST 2004 PING x.y.186.253 (152.110.186.253): 56 data bytes 64 bytes from x.y.186.253: icmp_seq=0 ttl=255 time=0.414 ms RR: x.y.186.253 x.y.186.253 x.y.186.3 --- x.y.186.253 ping statistics --- 1 packets transmitted, 1 packets received, 0% packet loss round-trip min/avg/max/stddev = 0.414/0.414/0.414/0.000 ms Thu Jun 24 22:46:58 SAST 2004 ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: Urgent 4.9 networking problems
The original ip 186.3 sets the broadcast - any aliases after that must have a /32 broadcast as they are aliases... That's correct isn't it (rest of list) ? Also - my connection is from 186.3 anyway, and those ip's are all functioning correctly.. I don't think that's where the problem is Thanks Dave -Original Message- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Kevin Stevens Sent: 24 June 2004 11:32 PM To: Dave Raven Cc: [EMAIL PROTECTED] Subject: Re: Urgent 4.9 networking problems On Thu, 24 Jun 2004, Dave Raven wrote: # ifconfig fxp1 fxp1: flags=8843UP,BROADCAST,RUNNING,SIMPLEX,MULTICAST mtu 1500 inet x.y.186.3 netmask 0xff00 broadcast x.y.186.255 inet x.y.186.1 netmask 0x broadcast x.y.186.1 inet x.y.186.15 netmask 0x broadcast x.y.186.15 inet x.y.186.14 netmask 0x broadcast x.y.186.14 inet x.y.186.142 netmask 0x broadcast x.y.186.142 inet x.y.186.33 netmask 0x broadcast x.y.186.33 inet x.y.186.124 netmask 0x broadcast x.y.186.124 inet x.y.186.250 netmask 0x broadcast x.y.186.250 inet x.y.186.122 netmask 0x broadcast x.y.186.122 inet x.y.186.25 netmask 0x broadcast x.y.186.25 inet x.y.186.127 netmask 0x broadcast x.y.186.127 I don't disagree with the other posters that mentioned DNS timeouts, but in addition those broadcast addresses aren't right. Since all the addresses are within the same /24 subnet, they should all be .255 (which is the default, so you wouldn't need to specify them. KeS ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED] ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
5.2.1 SMP problems
Hi all, Installed 5.2.1 on a high load server - albeit not recommended - and it was working perfectly with SMP support. It picked up 4 cpu's and appeared to be using them all. (The box is a dual 2.4gig Xeon machine). Anyway, the box was as I said under quite a bit of load, and would die about once a day - just reboot. I looked into it a bit and found whenever I put more load on (e.g. a make install in ports) it would panic - vm_load_page or something like that - sorry about being so vague. Anyway I took out SMP support, and now it boots with 2 cpu's - one normal Xeon + HT for the second (no idea how to disable that - but I don't need to)... its been running perfect for 4 days now and I put it under SERIOUS load.. has anyone else had SMP problems on 5.2.1? I couldn't find anything about it at first glimpse... Thanks Dave ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
RE: 5.2.1 SMP problems
Well all the tests I ran (about 2 memory testers and 2 benchmarks) caused the box to die - even the make installs did half the time. Now that I've disabled SMP there are no problems, and the same tests are working perfectly... The only thing I've changed is the SMP support, and now its been running for 4 days, before it wouldn't run for 12 hours Thanks Dave -Original Message- From: Albert Shih [mailto:[EMAIL PROTECTED] Sent: 04 June 2004 01:47 PM To: Dave Raven Cc: [EMAIL PROTECTED] Subject: Re: 5.2.1 SMP problems Le 04/06/2004 à 11:47:14+0200, Dave Raven a écrit Hi all, Installed 5.2.1 on a high load server - albeit not recommended - and it was working perfectly with SMP support. It picked up 4 cpu's and appeared to be using them all. (The box is a dual 2.4gig Xeon machine). Anyway, the box was as I said under quite a bit of load, and would die about once a day - just reboot. I looked into it a bit and found whenever I put more load on (e.g. a make install in ports) it would panic - vm_load_page or something like that - sorry about being so vague. Anyway I took out SMP support, and now it boots with 2 cpu's - one normal Xeon + HT for the second (no idea how to disable that - but I don't need to)... its been running perfect for 4 days now and I put it under SERIOUS load.. has anyone else had SMP problems on 5.2.1? I couldn't find anything about it at first glimpse... Have you shure this problem is with FreeBSD ? First check your memory with some software like memtest86 (http://www.memtest86.com/) When I've this kind of problem, always it's because some error in RAM. Regards -- Albert SHIH Universite de Paris 7 (Denis DIDEROT) U.F.R. de Mathematiques. Heure local/Local time: Fri Jun 4 13:43:52 CEST 2004 ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
mbuf cluster listing
Hi all, Is there anyway that I can list the proccesses using mbufs / mbuf clusters. I need to see which actual programs they are assigned to, not just what the usage is. Thanks Dave ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
mbuf clusters
Hi all, I have a FreeBSD box running a squid cache, FreeBSD 4.7-RELEASE-p9 i386, squid built from latest ports. (2.5-STABLE1). I'm not absolutely sure, but it appears that whenever I am running squid, the mbuf clusters on the bsd box rise at a steady rate, before reaching the limit, and the box's network failing to work. The box is under very small load and this is certainly not expected behaviour (I also have mbuf clusters set at 33280). I know this is not neccessarily a FreeBSD related question, but I am wondering how I can find out what it is that is using up the mbuf clusters without freeing them, as they steadily raise until they reach the max... How can I obtain more information about the mbuf cluster usage? (right now I'm just using netstat -m) Is there a way to force clearing them? Or to flush inactive usage, or anything? Quite lost here. Thanks Dave ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
Re: mbuf clusters
Thanks for the quick response, but I've looked into the options in the kernel, and to no avail. No matter what I set the mbuf clusters to, it just fills up. The larger the amount of clusters, the longer it takes, but it still fills up. Unless you meant options LIBMCHAIN #mbuf management library In which case I'm not really sure how that would help me as it seems to just be there to aid smb and nfs in some way (hard to find details). Could you perhaps explain this to me? Thanks again Dave - Original Message - From: Mantas S. [EMAIL PROTECTED] To: 'Dave Raven' [EMAIL PROTECTED] Sent: Wednesday, March 26, 2003 7:51 PM Subject: RE: mbuf clusters : grep mbuf /usr/src/sys/i386/conf/LINT : : :) : : -Original Message- : From: [EMAIL PROTECTED] : [mailto:[EMAIL PROTECTED] On Behalf Of Dave Raven : Sent: Wednesday, March 26, 2003 7:25 PM : To: [EMAIL PROTECTED] : Subject: mbuf clusters : : : Hi all, : : I have a FreeBSD box running a squid cache, : FreeBSD 4.7-RELEASE-p9 i386, squid built : from latest ports. (2.5-STABLE1). I'm not absolutely : sure, but it appears that whenever I am running squid, : the mbuf clusters on the bsd box rise at a steady rate, : before reaching the limit, and the box's network failing : to work. The box is under very small load and this : is certainly not expected behaviour (I also have mbuf : clusters set at 33280). I know this is not neccessarily : a FreeBSD related question, but I am wondering how : I can find out what it is that is using up the mbuf clusters : without freeing them, as they steadily raise until they : reach the max... How can I obtain more information : about the mbuf cluster usage? (right now I'm just : using netstat -m) Is there a way to force clearing them? : Or to flush inactive usage, or anything? : : Quite lost here. : : Thanks : Dave : : ___ : [EMAIL PROTECTED] mailing list : http://lists.freebsd.org/mailman/listinfo/freebsd-questions : To unsubscribe, send any mail to : [EMAIL PROTECTED] : : : ___ [EMAIL PROTECTED] mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-questions To unsubscribe, send any mail to [EMAIL PROTECTED]
hosts.allow
Hello all, This seems to be a fairly simple questions, but has been bothering me for a while now. I want to specify whole IP classes instead of single ips in my hosts.allow config file. These are the methods I have tried (unsuccessfully): sshd : 192.168.0.0/255.255.255.0 : allow sshd : 192.168.0. : allow sshd : 192.168.0/24 : allow sshd : 192.168.0.024 : allow What is the actuall way of doing this? Thanks, Dave. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
hosts.allow
Hello all, This seems to be a fairly simple questions, but has been bothering me for a while now. I want to specify whole IP classes instead of single ips in my hosts.allow config file. These are the methods I have tried (unsuccessfully): sshd : 192.168.0.0/255.255.255.0 : allow sshd : 192.168.0. : allow sshd : 192.168.0/24 : allow sshd : 192.168.0.024 : allow What is the actuall way of doing this? Thanks, Dave. To Unsubscribe: send mail to [EMAIL PROTECTED] with unsubscribe freebsd-questions in the body of the message
