IRC channel logs

<azert>for instance, I managed a disk less computer class where we decided to put all the storage in a raid server connected by fast Ethernet to a switch. Each client had their own / on nfs, but the were all sharing /usr in a single read only nfs export

<azert>I don’t know if it was a good idea, it was working real well for us

<azert>We had script to setup and configure a client in few minutes

<azert>With usrmerge this would have been more complex

<azert>Nowadays I’d probably opt for scripts copying the client os from the server to a local cheap nvme

<azert>But I’d frown upon the wasted fast storage and wasted replication possibilities..

<azert>I did some research about cluster distributed file systems back then, but didn’t find anything suitable

<gnu_srs>Who is interested to use topgit??

<ZhaoM>morning

<sneek>Welcome back ZhaoM, you have 1 message!

<sneek>ZhaoM, youpi says: I don't really understand your discussions. vim has integrated function-jumping support: :ta

<ZhaoM>youpi: the main reason is that sometimes ':ta' + ctags take me to the wrong definition (but I don't remember how to reproduce it)

<youpi>then :ts can allow to select a different one

<ZhaoM>so I'm wondering maybe setting up a lsp may be a better option

<ZhaoM>I guess the correctness is higher

<ZhaoM>ok I will try to use ':ts'

<ZhaoM>youpi: btw I quite interested in this line 'That wouldn't need any code modification if we were using an initial ramfs exposing that storeio on /dev/wd0s1.'

<ZhaoM>it's in the entry of the license conflict

<ZhaoM>in TODO list

<damo22>g]

<damo22>^ that also brings a dropdown for ctags

<ZhaoM>cool, I have used Ctrl+] for a long time but I have never realised I can get a dropdown :|

<azert>I think we really don’t have a good way to use an initrd, and my interpretation is that it is a grub shortcoming

<azert>In u-boot it would be straightforward to dump a file to memory and add a node in the device tree for it

<youpi>? in grub it's straightforward to load a module and let the kernel do whatever it wants with it

<youpi>it passes the address in the multiboot structure, and gnumach can do whatever it wants with it

<azert>Ideally gnumach would just provide a mechanism to pass this memory to the free pool once it has been utilized. Nothing else need to be done

<azert>by the kernel

<almuhs>hi. Do you know if it's possible to create a Hurd's translator in Rust?

<sneek>almuhs, you have 1 message!

<sneek>almuhs, damo22 says: yes, i have seen this error on all my intel thinkpads, but it is an improvement as before it did not probe the AHCI controller at all

<gnu_srs1>youpi: You mentioned I can create an amd64 image of Hurd using debootstrap/mmdebstrap: Can I supply my own built packages to the image?

<youpi>you can provide your own archive url on their command line yes

<gnu_srs1>Alternately: Can I install an amd64 image having usrmerge and then still use dpkg-fsys-usrunmess?

<youpi>I don't know about dpkg-fsys-usrunmess

<gnu_srs1>I'll contact the dpkg maintainer then.

IRC channel logs

2025-01-27.log