Greg
animals brewing
food and drink fiction
gardening general
health history
language music
multimedia opinion
photography politics
Stones Road house technology
Greg's diary
recent entries
Translate this page
Select day in January 2022:
Su Mo Tu We Th Fr Sa
1
2 3 4 5 6 7 8
9 10 11 12 13 14 15
16 17 18 19 20 21 22
23 24 25 26 27 28 29
30 31
Select month:
2021 May Jun Jul Aug
2021 Sep Oct Nov Dec
2022 Jan Feb Mar Apr
Today's diary entry
Diary index
About this diary
Greg's home page
Greg's photos
Network link stats
Greg's other links
Copyright information
    
Groogle

Sunday, 16 January 2022 Dereel Images for 16 January 2022
Top of page
next day
last day

Cleaning filters
Topic: photography Link here

How do some people keep their lenses so clean? I have bought a number of second-hand lenses, and the glass has always been spotless. I have given up on that a long time ago, and instead I put filters in front of the lenses. That protects the lens and makes it easier to clean, at least in theory.

Today I had a dirty filter. OK, take it off and clean it in the iso-propyl alcohol that I had bought for the purpose. It doesn't work! The smudges didn't go by themselves, and somewhere I ended up with granules of something that weren't there before. Lens cleaning tissue? It spreads the smudge around. In the end, things looked worse than before.

OK, to the bathroom. Wash the filter in warm water and kitchen detergent. Rinse copiously. Soak up the remaining droplets with a serviette (just touch, don't rub). Done!

No idea if this would work with hard water.


Upgrading teevee, day 3
Topic: technology, opinion Link here

It took me all day yesterday to get a boot image for my upgrade to teevee. And it was for a DVD. But I wanted to install from a memory stick. And in fact all I had was an SD card in a card reader. Can I even boot from that?

Yes! But that was the only good news. It failed to boot in exactly the same way as the memory stick did on Friday:


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20220116/big/ThinkCentre-boot-failure-1.jpeg
Image title: ThinkCentre boot failure 1          Dimensions:          4523 x 2884, 1575 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Sunday, 16 January 2022, thumbnails          All images taken on Sunday, 16 January 2022, small
Diary entry for Sunday, 16 January 2022 Complete exposure details

 

Clearly there's something wrong with the distribution media, since the computer itself (ThinkCentre 91p) works, and the boot media works with other ThinkCentres (M93p). What should I do?

The screen corruption at the top of the screen wasn't reproducible, but the hang was. I tried it with the standard boot, “safe mode”, boot to single user, and it always hangs in the same place. The -v option doesn't produce any additional information.

Entered a bug report. I don't have much hope of a quick fix, but this kind of problem needs to be documented.

So what do I do now? Copy an installation from another system? No, I don't think so. I'm trying to make this a textbook installation; I'll need to find another machine or other installation media. Another day with no progress.


Monday, 17 January 2022 Dereel Images for 17 January 2022
Top of page
previous day

More tiwi fun
Topic: technology, opinion Link here

OK, it's fairly clear that I can't boot FreeBSD 13 installation media on my spare ThinkCentre M91p. But it doesn't seem to be a problem on the M93p that I call dereel. Simple: remove the dereel disk from the M93p and put it in the M91p. That worked so easily that I wondered why I hadn't thought of it before. It also gives access to a better monitor: before I had to share with eureka's monitor 4 ():0.3).

OK, first thing after boot is to partition the new disk. How I hate this fake GUI interface that people have been running for decades! It evolved from the system that I described in the first edition of “Installing and running FreeBSD” in February 1996. The only difference is the disk partitioning menu: now it's even more GUIfied, and I decided to avoid it and escape to a shell to run gpart:


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20220117/big/tiwi-gpart.jpeg
Image title: tiwi gpart          Dimensions:          4547 x 2602, 1438 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Monday, 17 January 2022, thumbnails          All images taken on Monday, 17 January 2022, small
Diary entry for Monday, 17 January 2022 Complete exposure details

 

Where's ada0 there? It isn't. All three “geoms” are the same thing, the SD card, with different names. Confusingly, the card is almost exactly 1/1000 of the size of the disk: 7.4 GB instead of 7.3 TB. Miss that G instead of T and you can easily come to the wrong conclusion. I didn't find a way to access the disk. So back through the horrible menu after all, though I didn't create the /spool partition; I can do that later at the end, also giving me the opportunity to use custom newfs parameters.

Installation went without a hitch. Now install the ports.

root@tiwi:~ # pkg install bash emacs
The package management tool is not yet installed on your system.
Do you want to fetch and install it now? (y/N) y
Could not resolve host: pkg.freebsd.org
root@tiwi:~ #

Huh? What went wrong there? It works on freefall:

=== grog@freefall (/dev/pts/4) ~ 33 -> host pkg.freebsd.org
pkg.freebsd.org is an alias for pkgmir.geo.freebsd.org.

Local DNS issues? On freefall I had:

=== grog@freefall (/dev/pts/4) ~ 36 -> nslookup -q=soa freebsd.org
Server:         127.0.1.132
Address:        127.0.1.132#53

Non-authoritative answer:
freebsd.org
        origin = ns0.freebsd.org
        mail addr = hostmaster.freebsd.org
        serial = 2022011704
        refresh = 3600
        retry = 900
        expire = 1209600
        minimum = 3600

Authoritative answers can be found from:

=== grog@freefall (/dev/pts/4) ~ 37 -> date
Mon Jan 17 04:40:53 UTC 2022

That's the same as I got locally. But look at that serial number. The canonical YYYYMMDDvv. Dated today, when it's only 4:40, and there have already been four revisions! Clearly somebody is having problems, confirmed when I tried to update my git tree: it couldn't find git.freebsd.org either.

OK, wait...

=== root@dereel (/dev/pts/0) /usr/src-13 33 -> while :; do date; host git.freebsd.org; sleep 60; done
Mon 17 Jan 2022 15:48:04 AEDT
;; connection timed out; no servers could be reached

Mon 17 Jan 2022 15:49:14 AEDT
;; connection timed out; no servers could be reached

Mon 17 Jan 2022 15:50:24 AEDT
git.freebsd.org is an alias for gitmir.geo.freebsd.org.
gitmir.geo.freebsd.org has address 140.113.168.168
gitmir.geo.freebsd.org has IPv6 address 2001:f18:113:c5fb::e6a:1
gitmir.geo.freebsd.org mail is handled by 0 .
...
Mon 17 Jan 2022 15:52:27 AEDT
;; connection timed out; no servers could be reached

Mon 17 Jan 2022 15:53:37 AEDT
;; connection timed out; no servers could be reached

Mon 17 Jan 2022 15:54:48 AEDT
git.freebsd.org is an alias for gitmir.geo.freebsd.org.
...

OK, spend the time building a new world and kernel (13-STABLE). By the time I was done, freebsd.org was up to serial 2022011717.

When it had stabilized somewhat, finally got round to installing Xorg, bash and emacs. But it didn't quite finish downloading the 260 odd packages before things went away again.

But a retry helped, and by evening I had my three base ports installed. Now at least I have an xterm from tiwi on eureka:0.1, and another day has gone by.


How do you start an NFS server?
Topic: technology, opinion Link here

The source tree that I needed on tiwi is mounted on dereel. I'll need NFS to mount it.

I've been using BSD for not quite 30 years (since 19 March 1992), but there are still some basic things that I have never understood. One is: how do you start an NFS server on a running system? It's not enough to have an /etc/exports and a running mountd. Something about port mapping? rpcbind is your friend. nfsd? Yes, of course. Set all that up, but the mount still failed.

OK, use the only way I know: put this into /etc/rc.conf and reboot!

# NFS
nfs_client_enable=YES           # This host is an NFS client (or NO).
nfs_server_enable=YES           # This host is an NFS server (or NO).
rpc_lockd_enable="YES"          # Run NFS rpc.lockd needed for client/server.
rpc_statd_enable="YES"          # Run NFS rpc.statd needed for client/server.
rpcbind_enable="YES"

That worked, but I really don't understand. I'm pretty sure that rpcbind is essential, and that lockd isn't. statd maybe? In any case, the other big difference is that when I start nfsd from a shell, I only get one process. When I start it at boot, I get two. Here the related processes:

root     742   0.0  0.0  12828  2020  -  Ss   Mon11      0:00.05 /usr/sbin/rpcbind
root     785   0.0  0.0  15444  2152  -  Ss   Mon11      0:02.51 /usr/sbin/mountd -r -S
root     787   0.0  0.0  15316  1824  -  Is   Mon11      0:00.01 nfsd: master (nfsd)
root     788   0.0  0.0  12756  2288  -  S    Mon11      1:25.78 nfsd: server (nfsd)
root     791   0.0  0.0 277508  2104  -  Ss   Mon11      0:00.03 /usr/sbin/rpc.statd
root     794   0.0  0.0  15416  1976  -  Ss   Mon11      0:00.05 /usr/sbin/rpc.lockd

Another strange thing after rebooting: I got hundreds, no thousands, of old mail messages from dereel:

From MAILER-DAEMON Mon Jan 17 11:39:48 2022
Return-Path: <MAILER-DAEMON@dereel.lemis.com>
Received: from localhost (localhost)
        by dereel.lemis.com (8.16.1/8.15.2) id 20E7a12I001316;
        Fri, 14 Jan 2022 18:36:56 +1100 (AEDT)
        (envelope-from MAILER-DAEMON)
Date: Fri, 14 Jan 2022 18:36:56 +1100 (AEDT)
From: Mail Delivery Subsystem <MAILER-DAEMON>
Message-Id: <202201140736.20E7a12I001316@dereel.lemis.com>
To: grog@dereel.lemis.com
MIME-Version: 1.0
Content-Type: multipart/report; report-type=delivery-status;
        boundary="20E7a12I001316.1642145816/dereel.lemis.com"
Subject: Returned mail: see transcript for details
Auto-Submitted: auto-generated (failure)

[-- Attachment #1 --]
[-- Type: text/plain, Encoding: 7bit, Size: 0.5K --]

The original message was received at Sun, 9 Jan 2022 18:18:00 +1100 (AEDT)
from localhost [127.0.0.1]

   ----- The following addresses had permanent fatal errors -----
<grog@dereel.lemis.com>
    (reason: 450 4.2.0 /var/mail/grog: No locks available)

   ----- Transcript of session follows -----
>>> DATA
<<< 450 4.2.0 /var/mail/grog: No locks available
<grog@dereel.lemis.com>... Deferred: 450 4.2.0 /var/mail/grog: No locks available
Message could not be delivered for 5 days
Message will be deleted from queue

Why now? And how many of them are there?

=== root@dereel (/dev/pts/0) /usr/src-13 3 -> mailq
postqueue: warning: Mail system is down -- accessing queue directly
Mail queue is empty
=== root@dereel (/dev/pts/0) /usr/src-13 4 -> ps aux|grep mail
root    944   3.7  2.9 212516 173408  -  Ds   11:28    0:31.88 mail.local -l
root    912   2.0  0.1  18212   7784  -  S    11:28    0:18.24 sendmail: ./20E3C3aH064970
root    911   0.0  0.1  18084   6904  -  Ss   11:28    0:00.01 sendmail: accepting connec
smmsp   914   0.0  0.1  18084   6364  -  Is   11:28    0:00.00 sendmail: Queue runner@00:
root   1035   0.0  0.0  12864   2324  0  S+   11:40    0:00.00 grep mail
=== root@dereel (/dev/pts/0) /usr/src-13 22 -> sendmail -bp
postqueue: warning: Mail system is down -- accessing queue directly
Mail queue is empty
=== root@dereel (/dev/pts/0) /usr/src-13 23 -> /usr/sbin/sendmail -bp
/var/spool/mqueue is empty
                Total requests: 0

How can I be getting thousands of messages when the mail system is down? Is this another case of postfix masquerading as sendmail?

=== root@dereel (/dev/pts/0) /usr/src-13 24 -> wh sendmail
96117 -rwxr-xr-x  1 root  wheel  27032  5 Nov 04:54 /usr/local/sbin/sendmail
78659 lrwxr-xr-x  1 root  wheel     11 17 Jan 09:46 /usr/sbin/sendmail -> mailwrapper
=== root@dereel (/dev/pts/0) /usr/src-13 22 -> /usr/sbin/sendmail -bp
...
20C3W0Zl056277   961227 Wed Jan 12 14:32 <grog@dereel.lemis.com>
                 (Deferred: 450 4.2.0 /var/mail/grog: No locks available)
                                         <grog@dereel.lemis.com>
                Total requests: 1807

Yes! So it's running sendmail, and since postfix is installed, it hides the fact from the user.

/var/mail on dereel is a symlink to the directory on eureka, so it seems that I did need rpc.lockd to access it. But I've been running dereel for months, so while it's needed for some things, it's certainly not essential for everything. And now I had 7,700 messages to delete from root's inbox.


Checking up on Lorraine
Topic: general, opinion Link here

Lorraine Carranza has not been in the best of health lately, and today Helena Mirčić came by with horror stories of her being at death's door, taken to hospital with severe dementia, dead horses, and more.

Off in the afternoon, stopping first at Sally's, the neighbour down Westons Road who had looked after her. That was a good idea: we got a clearer answer from Sally. I think I've seen her before on 5 years ago when I found her dog on the main road. Yes, she had found Lorraine in her bed, half falling out and unaware of her surroundings. She screamed when touched—Sally says this is a symptom of dementia, though I didn't know it—and they had her taken off in an ambulance. She's somewhere in Ballarat, but she doesn't know where, and her son Claude is planning for a place for her in an aged care facility in South Melbourne. And yes, Lucy is dead, though she doesn't know what happened to the carcass. At the very least it seems that Lorraine didn't look after her, and since we had given Lucy to Lorraine, I'm a bit disappointed that we didn't hear anything about the problems. But she was old: Yvonne got her fully grown 15 years ago, so she must have been over 30.

Westons Road looks different. Eight years ago it looked like this (after I had managed to wedge the car across it):


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20140704/big/Wedged-car.jpeg
Image title: Wedged car          Dimensions:          4608 x 3456, 4976 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Friday, 4 July 2014, thumbnails          All images taken on Friday, 4 July 2014, small
Diary entry for Friday, 4 July 2014 Complete exposure details

 

Now it looks as if it wouldn't be possible:


https://lemis.nyc3.digitaloceanspaces.com/grog/Photos/20220117/big/Westons-Road.jpeg
Image title: Westons Road          Dimensions:          4651 x 3456, 9780 kB
Make a single page with this image Hide this image
Make this image a thumbnail Make thumbnails of all images on this page
Make this image small again Display small version of all images on this page
All images taken on Monday, 17 January 2022, thumbnails          All images taken on Monday, 17 January 2022, small
Diary entry for Monday, 17 January 2022 Complete exposure details

 


This page contains (roughly) yesterday's and today's entries. I have a horror of reverse chronological documents, so all my diary entries are chronological. This page normally contains the last two days, but if I fall behind it may contain more. You can find older entries in the archive. Note that I often update a diary entry a day or two after I write it.     Do you have a comment about something I have written? This is a diary, not a “blog”, and there is deliberately no provision for directly adding comments. It's also not a vehicle for third-party content. But I welcome feedback and try to reply to all messages I receive. See the diary overview for more details. If you do send me a message relating to something I have written, please indicate whether you'd prefer me not to mention your name. Otherwise I'll assume that it's OK to do so.


Greg's home page This month Greg's photos Greg's links

RSS 2.0 Valid XHTML 1.0!