December 2021 - Talk - lists.gtalug.org

war story: power failure makes CentOS system unbootable
by D. Hugh Redelmeier 30 Sep '22

30 Sep '22

tl;dr / spoiler: change fstab so that /boot/efi is not automatically mounted. See the recommendation at the end of this message. My gateway computer is a little PC running CentOS. It does not come back after a power failure. The reason (as best I can tell) is interesting and I think that I have a fix. My system is UEFI. It boots from a UEFI system partition, known as /boot/efi to Linux. If this gets corrupted, it won't boot. It is a VFAT partition. On my gateway (a Zotac Zbox RI321, a cute little box with two ethernet ports), the UEFI firmware apparently won't boot if the dirty bit is set in the system partition. CentOS normally runs with /boot/efi mounted. So when the system isn't shut down cleanly, the dirty bit will be on. Consequence: the system is not going to boot after a power failure. Odd observation 1: if I put a live Fedora 25 USB stick in the system and try to boot in this otherwise-unbootable state, CentOS boots from the main disk. So this really looks like a firmware bug/quirk. Odd observation 2: fsck doesn't seem to automatically fix the system partition. Once CentOS 7 booted, I dismounted /boot/efi and did an fsck on it. $ sudo fsck /dev/sda1 fsck from util-linux 2.23.2 fsck.fat 3.0.20 (12 Jun 2013) 0x25: Dirty bit is set. Fs was not properly unmounted and some data may be corrupt. 1) Remove dirty bit 2) No action ? 1 Leaving filesystem unchanged. /dev/sda1: 16 files, 2420/51145 clusters Googling got me to <https://www.centos.org/forums/viewtopic.php?t=50917> In particular, this advice seemed quite good: ==== recommendation ==== Two other changes I recommend for UEFI systems, to each OS's /etc/fstab. - For the /boot/efi mountpoint, add the mount options x-systemd.automount,noauto - Change fs_passno (last column) to a 1 or 2; the canonical fstab instructions suggest 2, but systemd treats 1 and 2 as the same. The first change means the EFI System partition will not be automatically read-write mounted at boot time; it's a bad idea this is the default because it puts the ESP at risk especially if ther are crashes, FAT has no journal and will therefore always be marked dirty in such a case; no other UEFI OS mounts the ESP by default. Second, if anything tries to access /boot/efi (read or write), systemd will automatically mount it, and because of the fs_passno 1 or 2, it will fsck it first and that fixes and clears the dirty bit in case it's set. Right now without these changes, it's just a matter of having the right number and bad timing of crashes to render the EFI System partition corrupt.

3 9

decent cheap ChromeOS tablet -- can run debian in a container
by D. Hugh Redelmeier 24 Jan '22

24 Jan '22

https://forums.redflagdeals.com/lenovo-canada-chromebook-10e-tablet-129-les… This is $129 tablet seems to be aimed at kids in schools. It's not particularly powerful, but it has a nice screen (1920x1200, 400nit), battery, and a solid build. This version of ChromeOS invites you to create a container with debian in it. The debian is from the debian repos: not customized. I've done so. I can run xterm, but it is awkward without a keyboard and mouse. I've loaded gnome desktop but I'm not sure how to start it. I wanted to use gnome's on-screen keyboard. Seems like a fun toy for this price. The weakest point is that it has only 32G of eMMC (disk) and no slot for an SD card. It has only one USB port and that is used for power. I was able to use a USB-C hub with "power delivery" to add more peripherals. The sale might end at any time. Perhaps the end of the day or the end of the weekend.

5 9

help with Rogerrs SMTP
by Ariel Burton 24 Dec '21

24 Dec '21

Hi everyone, I'm trying to get outgoing SMTP email to work with Rogers, and would appreciate any help. I am not looking for a full solution; all I need is for outgoing email to work from the command line in my user account. I don't need a system-wide solution, or a solution that works with GUI email readers. See below for a sample use-case. I used to have a solution that had lines like this in my $HOME/.mailrc: BEGINS set smtp=smtp.nl.rogers.com:587 set smtp-auth-user=MY-ROGERS-EMAIL-ADDRESS set smtp-auth-password=ROGERS-DEVICE-SPECIFIC-PASSWORD set replyto="Ariel Burton <MY-OTHER-EMAIL-ADDRESS>" set smtp-use-starttls set ssl-verify=ignore ENDS Something like this worked for many years, but stopped working a while ago. Earlier versions differed slightly, e.g., using "from" or "sender" in place of "replyto", or different port numbers. This is a transcript of how it fails: BEGINS /home/me> mail MY-OTHER-EMAIL-ADDRESS Subject: test test . EOT % smtp-server: 550 Request failed; Mailbox unavailable "/home/me/dead.letter" 12/365 . . . message not sent. ENDS Can anyone see what am I doing wrong, and how to fix it? Thanks, Ariel ==

6 6

"a web page is slowing down Firefox. To speed up your browser, stop that page."
by D. Hugh Redelmeier 22 Dec '21

22 Dec '21

I get this message when I read the Globe and Mail with Firefox on my Fedora 35 desktop. How can I figure out the cause?

4 4

LAN speed and USB ethernet dongles that go faster than 1 gigabit
by D. Hugh Redelmeier 19 Dec '21

19 Dec '21

2.5 and 5 gigabit USB ethernet dongles are available. They are not even too expensive. Does anyone have experience with these? Are they engineered in a way that doesn't put too much load on the host? Are they reliable? Background: I don't actually need anything faster than 1 gigabit yet, but the time to start migrating is always yesterday. Some ordinary motherboards come with 2.5 gigabit ports already. I don't know what the world will settle on as the new standard LAN speed. But it sure looks like we'll all want to be doing 2.5, 5, or 10 gigabits on our wired LANs in a few years. Does anyone have an inkling about what the next consensus will be? 2.5 gigabit hardware is cheap but might not be enough better than 1 gigabit to be worth the transition. So it might not become the consensus speed. 10 gigabit hardware is still quite expensive. More than I'm willing to pay right now. It is possible that 5 gigabits is a good compromise. Some example fast USB ethernet dongles: <https://www.servethehome.com/ugreen-usb-type-c-to-2-5gbe-cheap-network-adap…> This cost the authors about US$20 from Amazon but in Canada, right now, it is $50. The least expensive 2.5 that I found is generic, of unknown quality, for $33: <https://www.primecables.ca/p-390236-cab-pc-08159-usb-30-to-25-gigabit-lan-e…> Canada Computers has a few at $35: <https://www.canadacomputers.com/product_info.php?cPath=5_3598_3616&item_id=…> 5 gigabit is available, but more expensive. Here's a review (it contains links to reviews of other 5 gigabit units): <https://www.servethehome.com/sabrent-nt-ss5g-review-usb-to-5gbe-nic/> I don't see it on Amazon.ca even though it is on amazon.com.

1 0

meeting tonight?
by D. Hugh Redelmeier 15 Dec '21

15 Dec '21

I don't know if there will be a meeting tonight. That's up to the executive. But the lack of an email announcement either way puts it in doubt. The scheduled talk looks interesting: <https://gtalug.org/meeting/2021-12/> It doesn't say how to join Zoom or Google Meet (both are mentioned).

2 1

no meeting tonight
by D. Hugh Redelmeier 14 Dec '21

14 Dec '21

The meeting tonight is cancelled. We hope that we'll be better organized for the next meeting.

1 0

Booting linux from nvme disk?
by Peter King 10 Dec '21

10 Dec '21

Anyone had success with getting linux to boot from an nvme disk? I've been fighting with UEFI and Arch Linux all day now, trying to get a new computer to even recognize the nvme disk as a boot device. (Last time I encountered this problem I gave up and installed an ordinary HD to boot from.) For what it's worth, I have disabled Secure Boot, reformatted the nvme disk to have a new EFI partition without Windows, a swap partition, and a root partition; I'm trying to use efibootmgr (so no loading of a further bootloader), and, as far as I can tell, identified all the right partitions by device name or PARTUUID. Still no go. I even tried adding the parameter nvme_load=YES into the "root" part of the efibootmgr, and also adding nvme and vmd as modules in mkinitcpio.conf, all without any success. If necessary I'll just punt and install a regular HD to boot from, but that rather takes away from having an nvme disk in the first place. (Over the years I've learning to approach installing Linux with fear and loathing, with almost all the problems being with the bootloader -- from LILO through GRUB and GRUB2 now down to UEFI.) Any suggestions welcome! I didn't want to spend my weekend doing this. -- Peter King peter.king(a)utoronto.ca Department of Philosophy 170 St. George Street #521 The University of Toronto (416)-946-3170 ofc Toronto, ON M5R 2M8 CANADA http://individual.utoronto.ca/pking/ ========================================================================= GPG keyID 0x7587EC42 (2B14 A355 46BC 2A16 D0BC 36F5 1FE6 D32A 7587 EC42) gpg --keyserver pgp.mit.edu --recv-keys 7587EC42

10 14

Comparing XML/JSON files
by William Park 07 Dec '21

07 Dec '21

Hi all, I know of "xmldiff" and "jsondiff" to compare XML or JSON files. But, they don't allow me to exclude certain node or keyword. Do you know any command line program that does that, in one-shot? If not, then I have to resort to 1. For JSON, use 'jq' to remove keyword, and then compare. 2. For XML, use 'xmlstarlet' to remove node, and then compare. --William

4 5

a solved problem unsolved itself: WordPress, MySQL, UTF-8
by Stewart C. Russell 02 Dec '21

02 Dec '21

I have been running a WordPress blog hosted on a Linux-based shared host since WordPress became a thing. It has worked quite well from about 2004 up until a few weeks ago. Sadly, *something* recently decided my database encoding was wrong. And that something decided to "fix" it. It certainly "fixed it", but not in any way I could want. It also did the same for Catherine's blog. I know I didn't change any part of the config chain. As far as I can see: * the MySQL database still thinks the text is encoded in UTF-8; * Wordpress thinks the data is in UTF-8; * the web server is serving UTF-8. (NB: there's going to be some UTF-8 and hex chars in this message.) A typical post which shows this problem is https://scruss.com/blog/2016/02/27/t%c9%92k-b%c9%92ks-a-tiny-hardware-speec… When I should be seeing something like: [tɒk bɒks] 5b74 c992 6b20 62c9 926b 735d I'm seeing this in the page served up (and in the db text itself): [tÉ’k bÉ’ks] 5b74 c389 e280 996b 2062 c389 e280 996b 735d So the phonetic character U+0252 has been mangled into U+00C9 + U+2019. Every UTF-8 character seems to be affected this way. I wasn't expecting to wake up to a UTF-8 encoding problem this decade. There are a raft of "how to fix WP encoding issues" pages that show up in web searches, but the newest of them is from 2008 or so. I'm pretty much resigned to going through 16+ years of posts fixing this, but can mangled UTF-8 be recovered without rekeying? cheers, Stewart

6 11