Skip to main content

Upgrading busy systems

I’ve discovered a few inconvenient factors in upgrading a busy system — live.

Yes, it can be done.

But it’s not as simple as telling URPMI to upgrade everything it wants to, because then it swiftly consumes heaps of RAM (1.6GB, in this case) & then grinds to a halt as it starts to swap in search of even more RAM in order to sort out the competitive requirements of thousands of packages in one go.

My way around this has been to tell it to upgrade everything, then stop it before it does the actual upgrades, select a smallish chunk (maybe 50-100MB) of packages, & tell it to upgrade just that.

This works well until you select something which has some heavy requirements.

My bugbear here has been the X system, which includes the basic services, drivers for a zillion video chipsets, & countless X utilities which are dependent upon the version of X itself.

I have in mind a few key packages that it asks for as a part of this set of dependancies, and when I see one of those come up, I deduce which package(s) asked for it & try the chunk again without them.

This has the added advantage of interrupting only one or two services at a time, and that for short periods, whereas if it were to try upgrading the whole collection en-bloc, it might remove the old services early, & install their replacements late — or discover that a package it’s removing is internally dependent on some particular files etc belonging to another package, so remove the old one & not install a replacement.

With short install blocks, I get a detailed listing of the conflict, & to quickly identify the problem-child, remove it & its dependencies, & then “hand-install” a replacement. Also, I typically only lose one or two tools at a time; for example, vi unexpectedly expired for lack of a PERL library at once stage, so I simply used another editor while I found a useful replacement for the library.

Aaanyway... down to 2.55GB & 1420 packages now, & proceeding carefully.

Comments

ajdlinux said…
One of the reasons I dislike Mandriva is while RPM is pretty fast the frontend tools like urpmi and the control centre are *REALLY* slow.

APT can calculate an entire release upgrade in a few seconds - and it doesn't consume nearly as much RAM :) People dislike RPM for a reason...

Popular posts from this blog

every-application-is-part-of-a-toolkit at work

I have a LibreOffice Impress slideshow that I wish to turn into a narrated video. 1. export the slideshow as PNG images (if that is partially broken — as at now — at higher resolutions, Export Directly as PDF then use ‘pdftoppm’ (from the poppler-utils package) to do the same). 2. write a small C program (63 lines including comments) to display those images one at a time, writing a config file entry for Imagination (default transition: ‘cross fade’) based on when the image-viewer application (‘display,’ from the GraphicsMagick suite) is closed on each one; run that, read each image aloud, then close each image in turn. 3. run ‘Imagination’ over the config file to produce a silent MP4 video with the correct timings. 4. run ‘Audacity’ to record speech while using ‘SMPlayer’ to display the silent video, then export that recording as a WAV file. 4a. optionally, use ‘TiMIDIty’ to convert a non-copyright-encumbered MIDI tune to WAV, then import that and blend it with the speech (as a quiet b

new life for an old (FTX) PSU, improved life for one human

the LEDs on this 5m strip happen to emit light centred on a red that does unexpectedly helpful things to (and surprisingly deeply within) a human routinely exposed to it. it has been soldered to a Molex connector, plugged into a TFX power supply from a (retired: the MoBo is cactus) Small Form Factor PC, the assorted PSU connectors (and loose end from the strip) have been taped over. the LED strip cost $10.24 including postage, the rest cost $0, the PSU is running at 12½% of capacity, consumes less power than a laptop plug-pack despite running a fan. trial runs begin today.

boundaries

pushing the actual boundaries of the physical (not extremes, the boundaries themselves) can often remove barriers not otherwise perceived. one can then often resolve an issue itself, rather than merely stonewalling at the physical consequences of the issue.