Portál AbcLinuxu, 12. května 2024 18:02
3: eth1: BROADCAST,MULTICAST,UP mtu 1500 qdisc htb qlen 1000
link/ether 00:0e:0c:d0:1d:ef brd ff:ff:ff:ff:ff:ff
inet 192.168.101.1/23 brd 192.168.101.255 scope global eth1
inet nejaka.vereejna.ip.adresa/25 brd xxx.xxx.xxx.xxx scope global eth1
inet 10.0.0.1/24 brd 10.0.0.255 scope global eth1
4: eth1.2: BROADCAST,MULTICAST,UP mtu 1500 qdisc noqueue
link/ether 00:0e:0c:d0:1d:ef brd ff:ff:ff:ff:ff:ff
inet 10.1.2.1/24 brd 10.255.255.255 scope global eth1.2
2: eth0: BROADCAST,MULTICAST,UP mtu 1500 qdisc pfifo_fast qlen 1000
link/ether 00:11:09:92:1b:a7 brd ff:ff:ff:ff:ff:ff
inet 10.0.0.2/25 brd 10.0.1.255 scope global eth0
3: eth0.2: BROADCAST,MULTICAST,UP mtu 1500 qdisc noqueue
link/ether 00:11:09:92:1b:a7 brd ff:ff:ff:ff:ff:ff
inet 10.1.2.3/24 brd 10.255.255.255 scope global eth0.2
ping 10.1.2.1 -A
xxxxxxxxxxxxxxxxxxxxxxxxx
--- 10.1.2.1 ping statistics ---
77 packets transmitted, 69 received, 10% packet loss, time 1366ms
rtt min/avg/max/mdev = 0.121/455.918/1320.696/489.733 ms, pipe 42, ipg/ewma 17.9 79/1042.866 ms
) rozsvitili diody signalizujici link, ale sit nefungovala. Chovalo se to stejne v linuxu i Win2k3 serverech. Na radu zkusenejsiho kolegy jsme vymenili stavajici sitove kabely ( bezne kabely katogorie 5 ) za hodne drahe kategorie 6 ( jsou i mnohem silnejsi, asi lepe stinene apd. ) a vse funguje naprosto bez problemu. Jednalo se o stejne a pribuzne karty ( 1Gb Intel do cisco 3620 )
Muzes to zkusit, treba to pomuze. Pouzivame i 3com 1Gb NIC a ty tak citlive nejsou - fungujou i s obyc stavajicima kabelama k5.
, nejaky dalsi napad.... ?
lspci -v:
02:05.0 Ethernet controller: Intel Corp.: Unknown device 107c (rev 05)
Subsystem: Intel Corp.: Unknown device 1376
Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 11
Memory at f7e60000 (32-bit, non-prefetchable) [size=128K]
Memory at f7e40000 (32-bit, non-prefetchable) [size=128K]
I/O ports at c800 [size=64]
Expansion ROM at f7e20000 [disabled] [size=128K]
Capabilities: [dc] Power Management version 2
Capabilities: [e4] PCI-X non-bridge device.
Unknown device 107c
ale treba zbytecne ...
znervoznuje mev gentoo je prikaz update-pciids , kdyztak poslu co dela... nevim jak jinde...brachuv ntb s debianem zrovna nebeziUnknown device 107cale treba zbytecne ...
30.5.2007 06:52
koulinek | skóre: 19
| blog: koulonet
02:0d.0 Ethernet controller: Intel Corporation 82541PI Gigabit Ethernet Controller (rev 05)
Subsystem: Intel Corporation PRO/1000 GT Desktop Adapter
Flags: bus master, 66MHz, medium devsel, latency 32, IRQ 22
Memory at eb000000 (32-bit, non-prefetchable) [size=128K]
Memory at ea800000 (32-bit, non-prefetchable) [size=128K]
I/O ports at a400 [size=64]
[virtual] Expansion ROM at efe00000 [disabled] [size=128K]
Capabilities: [dc] Power Management version 2
Capabilities: [e4] PCI-X non-bridge device
Tak jestli nahodou ta karta neodchazi do vecnych lovist, obdobne se mi chovala jedna Atheroska.
...
Ethernet controller: Intel Corp.: Unknown device 107c (rev 05)
02:05.0 Ethernet controller: Intel Corporation 82541PI Gigabit Ethernet Controller (rev 05)
Subsystem: Intel Corporation PRO/1000 GT Desktop Adapter
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop- ParErr- Stepping- SERR- FastB2B-
Status: Cap+ 66MHz+ UDF- FastB2B- ParErr- DEVSEL=medium TAbort- Abort- MAbort- SERR- PERR-
Latency: 32 (63750ns min), Cache Line Size: 32 bytes
Interrupt: pin A routed to IRQ 11
Region 0: Memory at f7e60000 (32-bit, non-prefetchable) [size=128K]
Region 1: Memory at f7e40000 (32-bit, non-prefetchable) [size=128K]
Region 2: I/O ports at c800 [size=64]
Expansion ROM at f7e20000 [disabled] [size=128K]
Capabilities: [dc] Power Management version 2
Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0+,D1-,D2-,D3hot+,D3cold+)
Status: D0 PME-Enable- DSel=0 DScale=1 PME-
Capabilities: [e4] PCI-X non-bridge device
Command: DPERE- ERO+ RBC=512 OST=1
Status: Dev=00:00.0 64bit- 133MHz- SCD- USC- DC=simple DMMRBC=2048 DMOST=1 DMCRS=8 RSCEM- 266MHz- 533MHz-
Diky, upgradnul jsem pci-utils a ted se tvari jak ma, ale ! problemy pretrvavaji ..:(Ano, ten vypis (unknown device vs. spravne jmeno) lspci s vyse zminenymi problemy naprosto nesouvisi.
...Jen jak rikam jsou tam
dva kusy !! divne...
root@main:~/test2# cat /proc/interrupts
CPU0
0: 6109833 XT-PIC timer
1: 2 XT-PIC keyboard
2: 0 XT-PIC cascade
8: 1 XT-PIC rtc
10: 148961526 XT-PIC eth0
11: 174624348 XT-PIC eth1
14: 57446 XT-PIC ide0
NMI: 0
ERR: 0
root@main:~/test2# ip addr | grep ether
link/ether 00:0e:0c:d0:1d:f7 brd ff:ff:ff:ff:ff:ff
link/ether 00:0e:0c:d0:1d:ef brd ff:ff:ff:ff:ff:ff
link/ether 00:0e:0c:d0:1d:ef brd ff:ff:ff:ff:ff:ff
[petr@soban ~]$ cat /proc/interrupts
CPU0
0: 1683265 IO-APIC-edge timer
1: 863 IO-APIC-edge i8042
8: 1 IO-APIC-edge rtc
9: 0 IO-APIC-fasteoi acpi
10: 0 IO-APIC-edge MPU401 UART
12: 4 IO-APIC-edge i8042
14: 5093 IO-APIC-edge ide0
15: 33022 IO-APIC-edge ide1
16: 161048 IO-APIC-fasteoi nvidia
17: 72910 IO-APIC-fasteoi saa7133[0]
20: 10957 IO-APIC-fasteoi HDA Intel
21: 172809 IO-APIC-fasteoi eth0
22: 3 IO-APIC-fasteoi ehci_hcd:usb2
23: 26892 IO-APIC-fasteoi ohci_hcd:usb1
NMI: 333
LOC: 1683168
ERR: 0
Váš HW nemá podporu ACPI, APIC ?
after upgrading one of our routers last month, to AMD Athlon(tm) 64 X2 Dual Core Processor 4400+, we started to experience some weird problems with e1000 driver. Our previous CPU was Athlon 64 (not SMP). We're using Debian GNU/Linux stable 4.0 (etch), and vanilla kernel 2.6.21.1. The motherboard is Albatron K8NF4-AM2. Their site really sucks, and there is no BIOS updates at all. The router had 3 Intel Gigabith cards (82541PI chipsets according lspci) This router is increasing latency one-two times for few minutes to very high levels (sometimes to 3-4 seconds!). When the latency is up so high every few minutes and there is also packet loss in this time. This is happening for month or two maybe, and I tried so many kernels, from 2.6.19 to 2.6.21.1 and none of them are working OK. In the beginning there was also hardware reset of the adapters, but after some kernel version update this disappeared. I tried also every single driver from your project until now (including e1000-7.5.5), none of them resolved my problem. If I had to be accurate (from 2-3 releases of e1000 driver until now my problem with TX Hang (or RX Hang) disappear. But the latency and packet loss problem remains. You can see my ping to the router: .... Reply from X.X.X.X: bytes=32 time<1ms TTL=63 Reply from X.X.X.X: bytes=32 time<1ms TTL=63 Reply from X.X.X.X: bytes=32 time<1ms TTL=63 Reply from X.X.X.X: bytes=32 time=97ms TTL=63 Reply from X.X.X.X: bytes=32 time=124ms TTL=63 Reply from X.X.X.X: bytes=32 time=203ms TTL=63 Reply from X.X.X.X: bytes=32 time=276ms TTL=63 Reply from X.X.X.X: bytes=32 time=1062ms TTL=63 Reply from X.X.X.X: bytes=32 time=918ms TTL=63 Reply from X.X.X.X: bytes=32 time=183ms TTL=63 Reply from X.X.X.X: bytes=32 time=242ms TTL=63 Reply from X.X.X.X: bytes=32 time=209ms TTL=63 Reply from X.X.X.X: bytes=32 time=933ms TTL=63 Reply from X.X.X.X: bytes=32 time=514ms TTL=63 Reply from X.X.X.X: bytes=32 time<1ms TTL=63 Reply from X.X.X.X: bytes=32 time<1ms TTL=63 Reply from X.X.X.X: bytes=32 time=1ms TTL=63 Reply from X.X.X.X: bytes=32 time<1ms TTL=63 Reply from X.X.X.X: bytes=32 time=3ms TTL=63 .... What I've tried: - Different kernels (2.6.19-2.6.21.2) with native kernel driver, or with e1000 project latest driver) - Different e1000 driver versions (7.3.20-7.5.5) - Booting with "noapic" or acpi=off None of them helped in any way. The only workaround I found until now is very unacceptable - booting without SMP support - "nosmp" kernel option.
root@main:~/test2# uname -a
Linux main 2.4.29 #6 Thu Jan 20 16:30:37 PST 2005 i686 unknown unknown GNU/Linux
distro je Slackware 10.1 . Intel(R) Pentium(R) 4 CPU 3.20GHz 1024 Mb ram
30.5.2007 16:48
Luboš Doležel (Doli) | skóre: 98
| blog: Doliho blog
| Kladensko
30.5.2007 16:48
Luboš Doležel (Doli) | skóre: 98
| blog: Doliho blog
| Kladensko
30.5.2007 18:12
Luboš Doležel (Doli) | skóre: 98
| blog: Doliho blog
| Kladensko
Tiskni
Sdílej:
ISSN 1214-1267, (c) 1999-2007 Stickfish s.r.o.
30.5.2007 13:58