Aw: Re: [discuss] How much OpenIndiana is lagging behind Linux in live usb technology?

[Ignacio Soriano Hernandez <ignacio.soriano@me.com>]

[-- Attachment #1: Type: text/plain, Size: 2768 bytes --]

And the SoftwareManagementCenter python2 .. so not easy to digest as well .. As I wrote in a different posting .. there is something for everyone new joining the illumos/OpenIndiana/Tribblix (well if Peter allows :-)) world with skills and wanting to help :-) Though wireless would be lovely for those of us in a "mobile" world (and yes there are artifacts and I think even some old drivers in OI) :-) Cheers Iggi Peter Tribble <peter.tribble@gmail.com> schrieb am 30. Juli 2024 um 19:08: On Tue, Jul 30, 2024 at 12:00 AM nsgnkhibdk2cls0f via illumos-discuss < discuss@lists.illumos.org > wrote: Thank you for clarification. But the overall experience with OpenIndiana live usb is still very bad. You should have a look at it and do something to improve the performance. Historically, OpenSolaris/OpenIndiana/Solaris, based on the Caiman installer, have typically taken about twice as long as Ubuntu (which is probably a fair comparison out of the available Linux distros). I did some comparisons back in OpenSolaris, and I don't know whether there's been significant change in that ratio. Certainly the recent Debian installs I've done have been horribly slow. While OmniOS and Tribblix are very very much quicker, I'm not entirely sure how much of that work translates directly to OpenIndiana - with different targets in mind we can optimize in very different directions, and both those distros wrote something from scratch. A couple of the things I have in my old notes last time I looked at this (I haven't done a regular OpenIndiana install for some years now): 1. As the live environment is absolutely static, certain things could be precalculated ahead of time. I think it has to calculate the lists of files to transfer, which seems wasteful to do each time. 2. One useful trick I do in Tribblix, again based on the notion that you know precisely what the install environment looks like, is to pregenerate the full SMF respository (by doing a test boot and saving off the repository.db. it's going to be the same every time) so as to avoid waiting for SMF import at first boot, which can save a bit [especially in some virtualized environments where manifest import seems to run abnormally slowly]. But one of the snags is that there isn't a great deal of understanding of the live boot and installer internals. Being a good perl programmer many decades ago gave me Laziness, Impatience, and Hubris, and after spending a few minutes trying to work out what Caiman was doing I decided it was too complicated for me so I wrote all the tools for Tribblix from scratch in a fraction of the time. -- -Peter Tribble http://www.petertribble.co.uk/ - http://ptribble.blogspot.com/ illumos / illumos-discuss / see discussions + participants + delivery options Permalink

[-- Attachment #2.1: Type: text/html, Size: 4717 bytes --]