* [ISSUE] iproute2 bug
@ 2021-03-15 5:36 cryptarch
2021-03-15 5:39 ` iproute2 batch-mode fails to bring up veth cryptarch
` (5 more replies)
0 siblings, 6 replies; 7+ messages in thread
From: cryptarch @ 2021-03-15 5:36 UTC (permalink / raw)
To: ml
[-- Attachment #1: Type: text/plain, Size: 2581 bytes --]
New issue by cryptarch on void-packages repository
https://github.com/void-linux/void-packages/issues/29468
Description:
Hi,
(This should probably go upstream, but it's not clear whether iproute2 bug reports are welcome on the [netdev mailing list](http://vger.kernel.org/vger-lists.html#netdev). So I thought I'd start here because I first noticed the problem when pulling in iproute 5.11.0 on one of my Void Linux boxen today.)
### System
* xuname: `Void 5.10.23_1 x86_64 AuthenticAMD uptodate rrrmFFFFFFF`
* package: `iproute2-5.11.0_1`
### Expected behavior
Iproute2 batch mode should be able to build a working network namespace. See below in "how to reproduce" section for the example script `netns-test.sh`.
Expected usage and output (works in 5.10.0):
```
$ sudo ./bin/netns-test.sh
$ sudo ip netns exec netns-test ip route
default via 192.168.5.1 dev netns-test-1
192.168.0.0/24 via 192.168.5.1 dev netns-test-1
192.168.5.0/24 dev netns-test-1 proto kernel scope link src 192.168.5.2
192.168.5.1 dev netns-test-1 proto kernel scope link src 192.168.5.2
```
### Actual behavior
The inner veth doesn't come up properly:
```
$ sudo ./bin/netns-test.sh
$ sudo ip netns exec netns-test ip route
$ sudo ip netns exec netns-test ip link
Password:
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
39: netns-test-1@if40: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether 1a:96:4e:4f:84:31 brd ff:ff:ff:ff:ff:ff link-netnsid 0
```
### Steps to reproduce the behavior
Put this into a script such as `~/bin/netns-test.sh`:
```
#! /bin/bash
gw=192.168.5.1
ip=192.168.5.2
ns=netns-test
veth0=${ns}-0
veth1=${ns}-1
/usr/local/sbin/ip -b - << EOF
link add $veth0 type veth peer name $veth1
addr add $gw peer $ip dev $veth0
link set dev $veth0 up
netns add $ns
link set $veth1 netns $ns
netns exec $ns ip link set dev lo up
netns exec $ns ip link set dev $veth1 up
netns exec $ns ip addr add $ip/24 dev $veth1
netns exec $ns ip addr add $ip peer $gw dev $veth1
netns exec $ns ip route add default via $gw dev $veth1
netns exec $ns ip route add 192.168.0.0/24 via $gw dev $veth1
EOF
```
Run the script and then check whether it works with commands such as `ip netns exec netns-test ip link` and `ip netns exec netns-test ip route`.
Note that `git bisect` identifies `1d9a81b8c9f30f9f4abeb875998262f61bf10577` as the offending commit.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: iproute2 batch-mode fails to bring up veth
2021-03-15 5:36 [ISSUE] iproute2 bug cryptarch
@ 2021-03-15 5:39 ` cryptarch
2021-03-15 12:03 ` leahneukirchen
` (4 subsequent siblings)
5 siblings, 0 replies; 7+ messages in thread
From: cryptarch @ 2021-03-15 5:39 UTC (permalink / raw)
To: ml
[-- Attachment #1: Type: text/plain, Size: 391 bytes --]
New comment by cryptarch on void-packages repository
https://github.com/void-linux/void-packages/issues/29468#issuecomment-799127153
Comment:
Note that running the commands manually, one at a time, seems to work okay even in 5.11.0. Afaict, the problem _only_ affects batch mode, which makes sense, because the commit identified by git-bisect was purely focused on refactoring batch mode.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: iproute2 batch-mode fails to bring up veth
2021-03-15 5:36 [ISSUE] iproute2 bug cryptarch
2021-03-15 5:39 ` iproute2 batch-mode fails to bring up veth cryptarch
@ 2021-03-15 12:03 ` leahneukirchen
2021-03-15 12:03 ` leahneukirchen
` (3 subsequent siblings)
5 siblings, 0 replies; 7+ messages in thread
From: leahneukirchen @ 2021-03-15 12:03 UTC (permalink / raw)
To: ml
[-- Attachment #1: Type: text/plain, Size: 246 bytes --]
New comment by leahneukirchen on void-packages repository
https://github.com/void-linux/void-packages/issues/29468#issuecomment-799365028
Comment:
README.develop says to send bugs to netdev@, so please do that; I don't think this Void related.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: iproute2 batch-mode fails to bring up veth
2021-03-15 5:36 [ISSUE] iproute2 bug cryptarch
2021-03-15 5:39 ` iproute2 batch-mode fails to bring up veth cryptarch
2021-03-15 12:03 ` leahneukirchen
@ 2021-03-15 12:03 ` leahneukirchen
2021-03-15 18:11 ` cryptarch
` (2 subsequent siblings)
5 siblings, 0 replies; 7+ messages in thread
From: leahneukirchen @ 2021-03-15 12:03 UTC (permalink / raw)
To: ml
[-- Attachment #1: Type: text/plain, Size: 244 bytes --]
New comment by leahneukirchen on void-packages repository
https://github.com/void-linux/void-packages/issues/29468#issuecomment-799365028
Comment:
README.devel says to send bugs to netdev@, so please do that; I don't think this Void related.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: iproute2 batch-mode fails to bring up veth
2021-03-15 5:36 [ISSUE] iproute2 bug cryptarch
` (2 preceding siblings ...)
2021-03-15 12:03 ` leahneukirchen
@ 2021-03-15 18:11 ` cryptarch
2022-05-08 2:12 ` github-actions
2022-05-23 2:11 ` [ISSUE] [CLOSED] " github-actions
5 siblings, 0 replies; 7+ messages in thread
From: cryptarch @ 2021-03-15 18:11 UTC (permalink / raw)
To: ml
[-- Attachment #1: Type: text/plain, Size: 352 bytes --]
New comment by cryptarch on void-packages repository
https://github.com/void-linux/void-packages/issues/29468#issuecomment-799636231
Comment:
It seemed ambiguous:
> Most new features require a kernel and a utility component.
>
> Please submit both to the Linux networking mailing list
> <netdev@vger.kernel.org>
But you're probably right
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: iproute2 batch-mode fails to bring up veth
2021-03-15 5:36 [ISSUE] iproute2 bug cryptarch
` (3 preceding siblings ...)
2021-03-15 18:11 ` cryptarch
@ 2022-05-08 2:12 ` github-actions
2022-05-23 2:11 ` [ISSUE] [CLOSED] " github-actions
5 siblings, 0 replies; 7+ messages in thread
From: github-actions @ 2022-05-08 2:12 UTC (permalink / raw)
To: ml
[-- Attachment #1: Type: text/plain, Size: 293 bytes --]
New comment by github-actions[bot] on void-packages repository
https://github.com/void-linux/void-packages/issues/29468#issuecomment-1120334418
Comment:
Issues become stale 90 days after last activity and are closed 14 days after that. If this issue is still relevant bump it or assign it.
^ permalink raw reply [flat|nested] 7+ messages in thread
* Re: [ISSUE] [CLOSED] iproute2 batch-mode fails to bring up veth
2021-03-15 5:36 [ISSUE] iproute2 bug cryptarch
` (4 preceding siblings ...)
2022-05-08 2:12 ` github-actions
@ 2022-05-23 2:11 ` github-actions
5 siblings, 0 replies; 7+ messages in thread
From: github-actions @ 2022-05-23 2:11 UTC (permalink / raw)
To: ml
[-- Attachment #1: Type: text/plain, Size: 2584 bytes --]
Closed issue by cryptarch on void-packages repository
https://github.com/void-linux/void-packages/issues/29468
Description:
Hi,
(This should probably go upstream, but it's not clear whether iproute2 bug reports are welcome on the [netdev mailing list](http://vger.kernel.org/vger-lists.html#netdev). So I thought I'd start here because I first noticed the problem when pulling in iproute 5.11.0 on one of my Void Linux boxen today.)
### System
* xuname: `Void 5.10.23_1 x86_64 AuthenticAMD uptodate rrrmFFFFFFF`
* package: `iproute2-5.11.0_1`
### Expected behavior
Iproute2 batch mode should be able to build a working network namespace. See below in "how to reproduce" section for the example script `netns-test.sh`.
Expected usage and output (works in 5.10.0):
```
$ sudo ./bin/netns-test.sh
$ sudo ip netns exec netns-test ip route
default via 192.168.5.1 dev netns-test-1
192.168.0.0/24 via 192.168.5.1 dev netns-test-1
192.168.5.0/24 dev netns-test-1 proto kernel scope link src 192.168.5.2
192.168.5.1 dev netns-test-1 proto kernel scope link src 192.168.5.2
```
### Actual behavior
The inner veth doesn't come up properly:
```
$ sudo ./bin/netns-test.sh
$ sudo ip netns exec netns-test ip route
$ sudo ip netns exec netns-test ip link
Password:
1: lo: <LOOPBACK,UP,LOWER_UP> mtu 65536 qdisc noqueue state UNKNOWN mode DEFAULT group default qlen 1000
link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00
39: netns-test-1@if40: <BROADCAST,MULTICAST> mtu 1500 qdisc noop state DOWN mode DEFAULT group default qlen 1000
link/ether 1a:96:4e:4f:84:31 brd ff:ff:ff:ff:ff:ff link-netnsid 0
```
### Steps to reproduce the behavior
Put this into a script such as `~/bin/netns-test.sh`:
```
#! /bin/bash
gw=192.168.5.1
ip=192.168.5.2
ns=netns-test
veth0=${ns}-0
veth1=${ns}-1
/usr/local/sbin/ip -b - << EOF
link add $veth0 type veth peer name $veth1
addr add $gw peer $ip dev $veth0
link set dev $veth0 up
netns add $ns
link set $veth1 netns $ns
netns exec $ns ip link set dev lo up
netns exec $ns ip link set dev $veth1 up
netns exec $ns ip addr add $ip/24 dev $veth1
netns exec $ns ip addr add $ip peer $gw dev $veth1
netns exec $ns ip route add default via $gw dev $veth1
netns exec $ns ip route add 192.168.0.0/24 via $gw dev $veth1
EOF
```
Run the script and then check whether it works with commands such as `ip netns exec netns-test ip link` and `ip netns exec netns-test ip route`.
Note that `git bisect` identifies `1d9a81b8c9f30f9f4abeb875998262f61bf10577` as the offending commit.
^ permalink raw reply [flat|nested] 7+ messages in thread
end of thread, other threads:[~2022-05-23 2:11 UTC | newest]
Thread overview: 7+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-03-15 5:36 [ISSUE] iproute2 bug cryptarch
2021-03-15 5:39 ` iproute2 batch-mode fails to bring up veth cryptarch
2021-03-15 12:03 ` leahneukirchen
2021-03-15 12:03 ` leahneukirchen
2021-03-15 18:11 ` cryptarch
2022-05-08 2:12 ` github-actions
2022-05-23 2:11 ` [ISSUE] [CLOSED] " github-actions
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).