| MIF | Hi, I am having a networking issue, I cannot connect outside of the localhost on my ubuntu 20.04 server, this started to happen after I tried to install pi-hole and failed | 00:17 |
|---|---|---|
| MIF | I don't know how to fix it | 01:44 |
| derpadmin | MIF ping 8.8.8.8 does it work? yes? then it is just dns issue | 02:33 |
| MIF | yes it works | 02:39 |
| MIF | now that I know that is it a DNS issue, how do I fix it? | 02:39 |
| MIF | or let me resay that | 02:44 |
| MIF | How do I go about trying to find what is the issue and what is the best path to fix it? | 02:45 |
| === StathisA_ is now known as StathisA | ||
| dam_ | Hi, I am new to cloud-init and I was able to create an image with a / and /home partitions with nosource using subiquity. I would like to do that with cloud-init directly if possible. There is "fs_setup" but for now I do not know how I can use it create an image with / and /home. | 08:22 |
| dam_ | Is it possible to create an image with / and /home partitions from a ubuntu cloud image? | 08:24 |
| daftykins | Hey folks, i have used Landscape for some time with a free account limited to managing 5 hosts, but now i need something new to look after a new site with a few servers... i've been playing with cockpit today, any suggestions? I'm mostly just after pushing updates from a handy single spot. | 09:47 |
| === blueeagle is now known as BlueEagle | ||
| === dosaboy_ is now known as dosaboy | ||
| sdeziel | I have a setup where a given VM is connected to 2 networks each advertising IPv6 through SLAAC/RA. They both announce a default route and a machine connected to both picks the 'wrong' outgoing iface. Any ideas on what I could change on either networks to control which default route takes precedence? https://paste.ubuntu.com/p/TyWSpVGkWd/ | 14:31 |
| frickler | sdeziel: without checking in detail, I'd guess that "metric 100" wins over "metric 1024". you might be able to tune that in your router's RAs | 15:29 |
| sdeziel | frickler: yeah, dunno how tough. Those RAs metric 100 are coming out of dnsmasq and those with metric 1024 from radvd | 15:29 |
| sdeziel | any pointers? So far, I changed the preference to low for eth1 but as you said, the metric makes it win | 15:30 |
| === Ussat-1 is now known as Usssat | ||
| frickler | sdeziel: hmm, the metric seem to be actually generated by the kernel, it is not part of the RAs directly. default should be 1024, so I'm not sure where the 100 is coming from. on my local system I see 600 instead | 15:44 |
| frickler | the seems to be related by making it configurable, but doesn't explain where the old values are coming from https://lkml.org/lkml/2021/1/19/945 | 15:44 |
| === Usssat is now known as Ussat | ||
| sdeziel | the 100 metric is also what I see on the DHCP learned IPv4 routes... dunno, maybe it's systemd-networkd doing it? | 15:58 |
| frickler | sdeziel: ah, maybe dnsmasq is using dhcpv6 for extra config, including route metrics. can you paste a tcpdump or possibly radvdump of the RAs on your two interfaces? | 16:04 |
| frickler | also, do you have access to the dnsmasq config? there seem to be modes ra-only vs. ra-stateless, with the latter matching the above | 16:06 |
| sdeziel | frickler: dnsmasq conf: https://paste.ubuntu.com/p/dz8H87MmQS/ | 16:25 |
| sdeziel | frickler: the 2 pcaps: https://sdeziel.info/pub/t/ | 16:28 |
| sdeziel | frickler: I have access to every pieces on the network including the VM but I'd like to find a way to fix this routing priority by just changing the network infra instead of the VM's config | 16:56 |
| frickler | sdeziel: yes, that's understandable. I'll try to do some further testing myself, but not today. just for completeness, which kernel are you running and which ubuntu version? | 17:01 |
| sdeziel | frickler: 5.4 on 20.04 | 17:02 |
| sdeziel | frickler: I really appreciate the help, thanks! | 17:02 |
| memoryleak | Hi. I've difficulties trying to find a example for this: Install Ubuntu server with preseed file via virt-install using the iso file - is that possible? | 19:23 |
| rbasak | memoryleak: while it might still work, we generally left behind that antiquated method years ago and use cloud images now. | 19:53 |
| rbasak | It's really long winded to run an installer to set up a VM. We ship images that are ready for VMs. | 19:54 |
| rbasak | (unless testing the installer, etc) | 19:54 |
| pcfreak30 | Hello, how can I permanently increase the size of the /run tmpfs directory on bionic? | 20:36 |
| pcfreak30 | it is not defined in fstab but is in mtab | 20:36 |
| === Guest70304 is now known as MIF | ||
| === rfm_ is now known as rfm | ||
| sybariten | So... i installed ubuntu 20 (server, i believe) through a web panel from my VPS company. Is it likely that i get an autostarting apache with that per default? | 23:13 |
| sybariten | Cause i then instralled xampp, and realized that i seem to have two apache running .... maybe | 23:13 |
| sarnold | I could certainly see a VPS provider think they're being helpful by doing that, yeah | 23:16 |
| sarnold | I don't think our defaults would do that, but also don't know how to check :) | 23:16 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!