/srv/irclogs.ubuntu.com/2021/10/11/#lubuntu-devel.txt

guivercthanks kc2bez 00:59
guivercShould "### Notice about upgrading from previous versions of Lubuntu with LXQt:" be 'upgraded' to a "#" or larger heading  (or have a #Upgrading extra heading before it - to amplify message?)01:08
guiverca "." maybe should be before "New installations"  (but it may have been left off on purpose; i often do where it's after quoting & it can look weird with a space before it to avoid being treated as part of 'quoted' section01:11
guiverckc2bez, I've added a "most" in front of "other flavors use" (Installer) section b/c of Ubuntu Studio using calamares this release  (unless I'm wrong)01:14
lubot[telegram] <kc2bez> Those all sound reasonable to me. Feel free to edit or I can later.01:15
lubot[telegram] <kc2bez> I may move it to Discourse ahead of release as a Draft like the main Ubuntu release notes.01:16
guivercI won't add the "." ; only commenting here (I don't remember where ``. is a problem and where it's not.. but have been bitten before)01:18
lubot[telegram] <kc2bez> back ticks should work in both Discourse and WordPress (blog).01:19
guivercreading release notes... i wonder if we should create doc on discourse giving example of upgrading 512MB swapfile to something larger..01:19
guivercshould - my concern..  I've been bitten somewhere but can't recall where (moinmoin or wiki? maybe)01:20
lubot[telegram] <kc2bez> I can point swap to the wiki page we have on Discourse maybe.01:21
guivercLubuntu Manual ... maybe s/a Lubuntu Manual book/the Lubuntu Manual/01:21
lubot[telegram] <kc2bez> That works.01:22
guivercwe have 2 pages.. (one for 20.04 & 20.10, one for 21.04 & up); both mention increasing size only indirectly01:22
lubot[telegram] <kc2bez> I'll give it a look.01:23
* guiverc is still reading..01:24
guiverci'm going to make changes on 20.04 doco01:24
lubot[telegram] <kc2bez> No worries, take your time. I probably won't get back to it tonight.01:24
* guiverc added the "." too; it's showing a space in the preview (so is dealing with it correctly)01:26
guiverckc2bez, Rik & Thomas & apt-ghetto get mentioned in Extra Thanks, but only Thomas/apt-ghetto are mentioned in contributors.. Was this was intentional?01:33
* guiverc used those names in attempts to avoid pinging them01:34
guiverci've removed bold on "and" in Global Team01:35
lubot[telegram] <kc2bez> contributors = Lubuntu members.01:39
guivercack.01:39
guivercKGIII isn't one though (yet)01:40
lubot[telegram] <kc2bez> true. I can edit that section. It is mostly a carryover.01:40
guivercread thru complete; no mention of 1864787 or discourse; but I'm happy anywhere as I wanted it mentioned so I had somewhere to point support/users to & have discourse thread https://discourse.lubuntu.me/t/calamares-rare-failure-to-create-partition-mkfs-errors/277401:43
guivercs/anywhere/anyway ^01:44
kc2bezI think that needs to go towards the end in known issues. 01:50
lubot[telegram] <teward001> here's a stupid question: did we run any of those tests with debugging on to get the *exact* mkfs error underneath it?01:51
lubot[telegram] <teward001> because those're relevant.01:51
guivercwe didn't mention it in hirsute, but did so in focal & groovy (if I recall correctly)..  I only see 5-8 users report issues with it in a cycle (~6 months) & some are LTS users01:52
lubot[telegram] <teward001> was anyone here able to reproduce it?01:53
lubot[telegram] <teward001> or was it a reiser specific issue?01:53
guiverc@teward001,  not recently.. but in the past year & examined by tsimonq2  (no occurs with ext4, reiserfs, xfs, btrfs... - partition layout being busy is necessary)01:53
lubot[telegram] <teward001> cala mounting everything would be a trigger01:54
lubot[telegram] <teward001> because it never unmounts on an error case.01:54
guivercignore my 'year' ref sorry01:54
lubot[telegram] <teward001> no worries.01:54
guiverci can generally create it; count in last ~week is 3 fails & 1 clean on 1 box if I recall correctly01:55
lubot[telegram] <teward001> cool if you find reproduction steps that I can follow I'll generate a debug run01:55
guivercif i repeat the same steps on other boxes I won't get it.. Simon tried too, as I think Walter may have as well.. It's not always fail.. but I can create by (1) full disk install; (2) install alongside; then (3) manual install using / & /home (ie. loads of partitions on disk) SOMETIMES01:58
guivercs/as/and/ ^ 01:58
lubot[telegram] <teward001> i mean01:58
lubot[telegram] <teward001> i figured out how to force Cala to error hardcore mode01:58
lubot[telegram] <teward001> which is introduce a buggy NotImplementedError into the code randomly after mounting01:59
lubot[telegram] <teward001> causes a full explode and the partition table and partitions all to be busyh01:59
lubot[telegram] <teward001> pretty sure I can figure out a way to force explode it to reproduce the case. but i'll take a look in my breaks tomorrow01:59
guiverc@teward001, note: it's mostly me that encounters it, ie. older boxes using MBR partition tables (default for my [legacy] BIOS boxes)02:18
lubot[telegram] <teward001> that's the default my VM uses02:20
lubot[telegram] <teward001> unless I explicitly set UEFI under QEMU02:20
guiverc:)02:23
lubot[telegram] <teward001> guiverc: I can always force something to explode if I mount the directories first.  Cala doesn't do any unmounting until shutdown oddly enough so05:09
* guiverc ponders close() on shutdown.. next time calamares crashes, I'll endeavour to restart; exit then restart & see if i get the misbehaving I get when I have that error...05:33
lubot[telegram] <teward001> guiverc: I assume this was replicated on pretty much every live CD between last LTS and today's daily.  because I have a 21.04 ISO I downloaded to help someone yesterday and if it's replicateable there then I don't need to download the daily.14:33
-queuebot:#lubuntu-devel- Builds: Lubuntu Desktop amd64 [Impish Final] (20211011) has been added17:48
guiverc@teward001, if you're asking about 1864787 (failure to create partition)... I don't know, I've had dailies where I've installed twice & had no issues.. but next day it's back to re-occurring.  The issue doesn't always appear, but has seem easier on some dailies than others (but that could just be the doesn't always occur/randomness)22:09

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!