Bug 15026
| Summary: | Сеть в контейнере доступна при существующем "снаружи" IP | ||
|---|---|---|---|
| Product: | Branch 4.0 | Reporter: | Konstantin Pavlov <thresh> |
| Component: | vzctl | Assignee: | Dmitry V. Levin <ldv> |
| Status: | CLOSED FIXED | QA Contact: | Q.A. 4.0 <qa-4.0> |
| Severity: | normal | ||
| Priority: | P2 | CC: | aspsk, boris, lakostis, ldv, mike, shrek, vsu |
| Version: | 4.0 | ||
| Hardware: | all | ||
| OS: | Linux | ||
| Bug Depends on: | |||
| Bug Blocks: | 14167 | ||
Это не сизиф. Предлагаю по возможности проверить на текущем ovz-el. <thresh> gvy: Миша, ничего там проверять не надо, я это всё давно исправил: http://git.openvz.org/?p=vzctl;a=commit;h=c8c2e8caae6481546c9ad81df945822cd0ec5d89 Соответственно, это было исправлено в vzctl >= 3.0.24-alt1. |
Description
Konstantin Pavlov 2008-03-21 14:53:59 MSKkernel-image-ovz-smp-2.6.18-alt22 [root@office-vz1 ~]# vzctl start 98 Starting VE ... VE is mounted Adding IP address(es): 10.192.168.98 arpsend: 10.192.168.98 is detected on another computer : 00:19:b9:1f:b1:e4 vps-net_add WARNING: arpsend -c 1 -w 1 -D -e 10.192.168.98 eth0 FAILED Setting CPU units: 1000 Configure meminfo: 131072 Set hostname: vps98.immo VE start in progress... [root@office-vz1 ~]# vzctl enter 98 entered into VE 98 [root@vps98 /]# ip ad 1: lo: <LOOPBACK,UP,10000> mtu 16436 qdisc noqueue link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 inet 127.0.0.1/8 brd 127.255.255.255 scope host lo 3: venet0: <BROADCAST,POINTOPOINT,NOARP,UP,10000> mtu 1500 qdisc noqueue link/void inet 127.0.0.1/24 scope host venet0 inet 10.192.168.98/32 scope global venet0:0 [root@vps98 /]# ping ya.ru PING ya.ru (213.180.204.8) 56(84) bytes of data. 64 bytes from ya.ru (213.180.204.8): icmp_seq=1 ttl=56 time=103 ms 64 bytes from ya.ru (213.180.204.8): icmp_seq=2 ttl=56 time=126 ms 64 bytes from ya.ru (213.180.204.8): icmp_seq=3 ttl=56 time=93.1 ms 64 bytes from ya.ru (213.180.204.8): icmp_seq=4 ttl=56 time=36.4 ms При этом на HN: [root@office-vz1 ~]# tcpdump -eni eth0 icmp tcpdump: verbose output suppressed, use -v or -vv for full protocol decode listening on eth0, link-type EN10MB (Ethernet), capture size 96 bytes 19:16:27.959620 00:03:47:f1:54:95 > 00:09:6b:63:76:11, ethertype IPv4 (0x0800), length 98: 10.192.168.98 > 213.180.204.8: ICMP echo request, id 36644, seq 1, length 64 19:16:27.969843 00:09:6b:63:76:11 > 00:03:47:f1:54:95, ethertype IPv4 (0x0800), length 98: 213.180.204.8 > 10.192.168.98: ICMP echo reply, id 36644, seq 1, length 64 19:16:28.961465 00:03:47:f1:54:95 > 00:09:6b:63:76:11, ethertype IPv4 (0x0800), length 98: 10.192.168.98 > 213.180.204.8: ICMP echo request, id 36644, seq 2, length 64 19:16:28.986646 00:09:6b:63:76:11 > 00:03:47:f1:54:95, ethertype IPv4 (0x0800), length 98: 213.180.204.8 > 10.192.168.98: ICMP echo reply, id 36644, seq 2, length 64 19:16:29.961441 00:03:47:f1:54:95 > 00:09:6b:63:76:11, ethertype IPv4 (0x0800), length 98: 10.192.168.98 > 213.180.204.8: ICMP echo request, id 36644, seq 3, length 64 [root@office-vz1 ~]# ip li sh eth0 4: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:03:47:f1:54:95 brd ff:ff:ff:ff:ff:ff На той машине, где висит адрес, назначенный VPS'у: [root@snowwhite ~]# ip ad sh eth0 2: eth0: <BROADCAST,MULTICAST,UP,LOWER_UP> mtu 1500 qdisc pfifo_fast qlen 1000 link/ether 00:19:b9:1f:b1:e4 brd ff:ff:ff:ff:ff:ff inet 10.192.168.97/22 brd 10.192.171.255 scope global eth0 inet 10.192.168.98/22 scope global secondary eth0 в dmesg ничего нет (широковещательный arpsend не срабатывает) однако на default gateway arpwatch замечает смену мак-адреса для 10.192.168.98. При откате на -alt18 проблема решается.