Show Idle (>14 d.) Chans

← 2020-05-24 | 2020-05-26 →
feedbot: << whaack -- Guitar Practice Log 6
feedbot: << whaack -- Guitar Practice Log 5 - Pepe Romero Stretches and Basic Shoulder Anatomy
jfw: diana_coman: a review by tonight was probably doomed at the point of not quite finishing things off before checking out on Thursday.
jfw: anyway, had a nice day of clearing pine needles from gutters, sawing off boughs that were impinging on suspended cables or otherwise in the way, and a bit of relaxing.
jfw: trinque: last I heard, we have verbal agreement to sign from two prospective clients - one for our beginner and intermediate training, and one for consulting on a new corporate IT deployment.
jfw: hi _joerodgers, you might need to stick with the Joe65 name for now to stay voiced (if you stick around you'll need to get a GPG key registered). this is a learning place focused on practical & sustainable growth led by diana_coman; there are indeed people here interested in & knowlegeable about Bitcoin among other things, and quite a bit of history - first of all in the logs linked from the
jfw: topic, and second from TMSR from which this started as a branch.
bvt: diana_coman: i don't have specific numbers, i'd need to test it on something like mp-wp to get them. i did not get noticeable slowdown with linux kernel and vtools, but i did not measure anything, just checked how 'fast' it feels. that's something i intend to do for the writeup. tbh back in the day i barfed looking at the code already, so decided to can it until someone comes up with a real tree with this
bvt: problem.
diana_coman: bvt: heh, that barfed looking at the code already sounds to me measurement enough to give it some more thinking at the very least before pushing it as part of the main trunk; and the more I think about it, the more it seems to me that the main problem is for now one of better error message first of all rather than necessarily one of "should press it anyway"; I can still make the case for both options but I can also make the case ...
diana_coman: ... *against* both options so I'd rather go for the smallest and least costly (from all povs) change tbh - can it spit the name of the file + vpatches involved instead of just shouting "loop"? and to mark this at vtree as well instead of just silently doing nothing?
diana_coman: jfw: take up gardening! heh
diana_coman: _joerodgers: as jfw said really, there's quite a lot to both logs and, especially, context to make sense of it all; it tends to help/work easier for you if you actively engage though so maybe come around to talk one day from 7pm UTC when I'm more usually around; if you don't have voice, just ask for it in pm (me or someone else who is voiced)
sonofawitch: 2020-05-25 07:20:47 (#ossasepia) jfw: hi _joerodgers, you might need to stick with the Joe65 name for now to stay voiced (if you stick around you'll need to get a GPG key registered). this is a learning place focused on practical & sustainable growth led by diana_coman; there are indeed people here interested in & knowlegeable about Bitcoin among other things, and quite a bit of history - first of all in the logs linked from the
diana_coman: cruciform: how's the moving going?
diana_coman: whaack: reports seem to work not-bad for guitar practice at least; are you sure you don't want to do anything else at all?
bvt: if i liked that code, i would have just released the vpatch immediately with performance benchmark, so i agree with 'case against'. i will look into improving the error reporting code -- this should be possible (with slightly more code in, though the precise loop analysis would take approximately the same amount of ada code as went into that experimental vpatch.
diana_coman: bvt: so maybe not go directly for precise loop analysis then - what are the less-costly-but-still-better-error-message options there?
diana_coman: I guess at a stretch, since the offending vpatches are identified, one will just script the "find the matching hashes", not like it's a big deal; only I'd rather not have it on the side, if it can be avoided
diana_coman: joerodgers uhm, just leave it connected somewhere, it's not a problem.
diana_coman: !!up #ossasepia joerodgers
deedbot: joerodgers voiced for 30 minutes.
diana_coman: !!key joerodgers
deedbot: Not registered.
diana_coman: joerodgers: if you register a key with deedbot, I'll rate you and you'll then be able to voice yourself.
joerodgers: I will gladly do that, sorry for joining and exiting, just got an IRC client setup and channels saved. I will track down how to register with deedbot now
diana_coman: !!help
diana_coman: joerodgers: do you know what the WoT is?
joerodgers: I see it is a trusted group of people from TMSR, but that assumption mightbe wrong
diana_coman: heh, don't take words to mean what a first glance might suggest, it's generally not a very good strategy anyway; the canonical ref for the WoT is the article on
feedbot: << Young Hands Club -- RMD w27-29 review, May 4-24th, 2020
diana_coman: joerodgers: see deedbot's help page, you need to use that !!register command
joerodgers: thank you, I am reading the docs now
diana_coman: bots answer each to their own prefix, not just randomly to any text
jfw: diana_coman: indeed I could even take up gardening now
feedbot: << Young Hands Club -- RMD w30 plan, May 25-29th, 2020
diana_coman: jfw: so then - plan sorted: take up gardening!
diana_coman: dorion: yo, welcome back but *slow down"! lolz
deedbot: A8A2746C266D1447AAD7A0B0B087C5F744EE8888 registered as joerodgers.
diana_coman: congrats joerodgers
joerodgers: thanks for the guidance
diana_coman: no problem at all
dorion: diana_coman thanks. where looks to you to be the best place for me to slow down ?
dorion: joerodgers, welcome.
diana_coman: dorion: lol, at writing, before publishing; that has got to be the shortest possible review of 2 weeks worth of time that I ever saw !
diana_coman: but hm, now looking at it, wtf happened to the categories on yh
diana_coman: this is a new one
diana_coman: did anyone notice -when did the categories there vanish?
dorion: diana_coman I noticed the change today, thought it was intentional.
diana_coman: neah, I have to bring them back; only now I wonder just when did I manage to nuke them.
jfw: I did not notice that as of my last (May 18)
diana_coman: to make it even weirded, they *are* in the db, ugh.
dorion: diana_coman I see what you mean now with slowing down, I'll add to it.
diana_coman: billymg: do you know where exactly does the UI get the categories /what does it need other than the stuff in the db?
diana_coman really does not want to go about comparing the backups, ugh
jfw: - ref I was looking for on gardening: I'd just be missing out on the miserable weather now
diana_coman: huh, what the issue was: apparently the UI relies on a temporary file to gather the list of categories to show; and the VM was low on disk space (unrelated - I ran some other stuff there) and so...this nuked the categories from showing...
diana_coman: anyways, space is now plenty, should not run out of it again either, so categories are showing again all fine
diana_coman: ftr, nothing was lost/wrong at any time in the db or anything - it was at all times a matter of display, but ugh.
billymg: diana_coman: this is for categories on the public-facing www, or in the admin panel?
diana_coman: billymg: both
diana_coman: they were not showing anywhere, neither in the admin panel, nor in the contributor/writer (ie non-admin user) panel, nor on the www (all articles were shown as Uncategorized)
billymg: diana_coman: what was the temp file called? or did you find where it's created/referenced (i wouldn't mind taking a look just for my own understanding of it)
diana_coman: funnily enough (and the ~only clue I had that something funny was going on) was that the whole interface otherwise correctly named the *default* category in other places (eg where it warns that deleting a category does not delete the articles)
diana_coman: billymg: I didn't dig that deep so unfortunately I can't help you there much; basically I checked everything in the db including that all had the correct parents etc; I checked that it was even as it had been before; then I noted that discrepancy; so looked around at what *else* could be the trouble
diana_coman: at any rate, I think it's a lousy way this is done and I can't quite understand why would it be done differently than all the rest or wtf does it need now disk space for
billymg: diana_coman: hrm, seems bizarre indeed
diana_coman: it's true that I never managed to run out of space on a blog-running machine but then again, this is kind of a special-situation-machine by all measures.
billymg: definitely seems strange that the lits would be written to a temp file and not just in memory
billymg: list*
diana_coman has always had trouble with categories & wordpress
diana_coman: I don't know why did they mess them up so much
billymg: diana_coman: how was the ram utilization doing on the box?
diana_coman: billymg: uhm, I didn't check specifically; there was though nothing special running at the time (the script that had gotten the drive full had finished anyway)
billymg needs to look closer at this in the code, but at the moment is focused on just ripping out all the js cruft
diana_coman: billymg: no rush with this as such; it was a wtf-moment but admiteddly, out-of-disk-space is not an usual state anyway and generally gets sorted before it causes this sort of wtf.
diana_coman: !!v 76F4168158928BD1BA9D369F55A353572E71255DA919C7B6AB7C848F1DAC367F
deedbot: diana_coman rated joerodgers 1 << new to it all
diana_coman: joerodgers: there you go; you should be able to voice yourself now, see deedbot's help page
billymg: diana_coman: ok got it
feedbot: << Trilema -- Nuts, absolutely fucking nuts.
← 2020-05-24 | 2020-05-26 →