Looks good first glance to me. Next rolling will have it included.
Can you please test installing this package and feedback to me?
@pvelati thanks for the update!
hi @c-po , I've tried installing the 3.8.3 version but it's not compatible with the new CloudFlare API v4. The only way to make ddclient working with cloudflare is using the 3.9.0.
Even better, looks like wimpunk is maintaining again and there is a fresh release https://github.com/ddclient/ddclient/releases 3.9.0
Debian Buster uses 3.8.3 https://packages.debian.org/buster/ddclient which could be considered as using a proper source tree.
@dmbaturin bump? I'd like to submit a PR for this issue, but would like some guidance regarded to my comment above.
Sat, Oct 19
Okay, thanks for the info.
It is true what @Viacheslav say, there is only possible to run one instance of bgp om a given router.. when using vrf's the bgp running inside the vrf is a subset of the main instance. to confirm, start vtysh and try to create multiple processes.. it will likly fail :)
We are a little confused.
As far as I know, Bgp instance on all router's platforms always one.
And it use socket with 179 tcp port.
VRF is working on same bgp instance, but support multiple autonomous system at once.
Agree with you, thank you. We need fixed these moments.
Unfortunately it doesn't work without gw-ip-address also in [chap-secrets]. In my early trials I've tried that, and the clients received IP's from the pool instead the static values.
I think better calculate gw-ip-address automatically. outside-nexthop excluded from required for migration reasons.
Not necessary define gw-ip-address in [chap-secrets] section, it works without any issue if defined it in [ip-pool] only.
@Dmitry Thanks for pointing that out, I've looked on the wiki (https://wiki.vyos.net/) and searched for a description for outside-nexthop, and failed to find it, and the in-line help while configuring doesn't help much.
You can set outside-nexthop which fixed this. Also I think we can calculate first ip for gw-ip-address if outside-nexthop is not defined by cli.
@runar at the current time this is true only one BGP instance is supported. But when we incorporate VRF one day which is definately on the roadmap this could change.
Verified in VyOS 1.2-rolling-201910190658
As for my understanding frr only supports a single bgp instance running at a time. But i've not verified thid completly.
ISO has been pulled. Next rolling will have the fix.
What about upcoming VRF when we possibly have multiple bgp processes?
This is already fixed in vyos-1x package but there is currently a problem with the CI.
Out of curiosity, I tried 201910170117, and it was fine (at least for booting). The 20191018 image may be broken - should it be pulled?
This is present in branches crux and current too, so adding tag for crux.
Via the process of elimination I found the culprit - comments with special characters in them, that at some point in the upgrade cycle got converted into backslash-escped octal:
description "\197\160" # this was once Š description "\196\140 \196\141" # Č č
So old versions allowed these characters in comments and worked fine with them, then at one opint a upgrade converted them to backslash octal numbers, which still worked until at some point they started causing the error.
This works as expected
I'm still having an issue with using build-ami to create an AMI in us-gov-west-1.
Thanks for your contribution.
Fri, Oct 18
Perfect. Thanks a lot guys.
As the AS number is a bgp specific attribute i don't think it's wise to move it out of the bgp hierarchy .
@c-po I can't see the issue here, that seems like normal expected behavior (tabbing gives a list of all possible nodes, even the long ones?)
I agree that if there is a problem, it's for a separate issue and describe what the problem and expected behavior is in more detail there.
fixed in 1.2.3
@c-po what you are describing don't look like a bug.. its more like a feature :) anyways its not related to this issue so please open another ticket on this.
I see no reason to add wget as curl is superior. We should set this ro Wontfix
@Harliff Does curl work for you?
Is that still and issue, sorry I lost track a little while I was busy with other stuff.
already added to the ocumentation: https://vyos.readthedocs.io/en/latest/system/proxy.html
good first implementation thx. cna you please also update docs?
I have an idea, I can either write it to profile.d, that is exporting http_proxy, https_proxy and ftp_proxy into the shell env, and in the install-image script if the profile files exists, I load it which exposes these variables as well and curl is working with no issue. If removed, that file won't exists and curl works like it did before. If the proxy variables shouldn't be in the user environment, I can write it to a particular file only used by scripts which which would need that information.
curl only accepts ~/.curlrc, so that can become a hassle with multiple home directories on a box.
We only have curl and wget for outside communication so this should be fine, whereas curl is the preferred way - thus this is more then fine and can be altered if its really required. Keep it simple.
Root cause identified and fixed. Please test @albeu.
That would work but it's only for a single programm you define it. I think it could be enough for the beginning. I still have to check if curlrc is being read when invoked from the perl script, it usually should.
Why bot simply create /etc/wgetrc? Can curl read that file? If not, /etc/curlrc.
Thu, Oct 17
The removal makes a little headache. Setting it system wide is not an issue at all, writing and execute in profile.d. Removing it would require to logout and login again to re-read the bash.profile. I may have to rethink that.