Bug#1043145: Upgraded ntpsec to 1.2.2+dfsg1-1+deb12u1: fewer servers remain

2023-08-07 Thread Richard Laager 
The "pool" directive spins up more associations than it needs and then 
prunes them back (which takes a while).


https://docs.ntpsec.org/latest/quick.html#pool
https://docs.ntpsec.org/latest/discover.html#assoc

With 4 pool servers each returning 4 A records and one of them returning 
4  records, it seems to me you could end up with potentially 20 
servers initially, which will then be pruned back to 7 (per tos maxclock*).


* Note that tos maxclock is 11 in the Debian default, as pool entries 
themselves count. I personally think that's wrong, but changing it 
creates a compatibility concern.


--
Richard



OpenPGP_signature
Description: OpenPGP digital signature

Bug#1043145: Upgraded ntpsec to 1.2.2+dfsg1-1+deb12u1: fewer servers remain

2023-08-06 Thread Thure Dührsen 
Dear Maintainer,

after upgrading ntpsec to 1.2.2+dfsg1-1+deb12u1, the behaviour is slightly 
different:

only ten servers remain after disconnecting from the internet,

[tdu:debiantutor] ~
 date; ntpq -pn | grep -Fv pool
Sun  6 Aug 21:02:52 CEST 2023
 remote   refid  st t when poll reach   delay   offset   jitter
===
+78.47.168.188   17.253.14.1252 u 1103  1280  46.9383  -4.3886 124.1287
+144.76.0.164193.190.230.65   2 u 1101  1280  33.9219  -7.6402 217.2358
-84.46.245.73237.17.204.952 u 1107  1280  45.2009 -13.4787 120.2817
-54.38.156.2210.1.1.2 2 u 1109  1280  29.9127  -7.8516  43.2926
+162.159.200.123 10.124.8.190 3 u 1102  1280  32.4660  36.9115 120.6871
+86.90.148.127   125.199.53.762 u 1105  1280  42.9345  37.2525  33.6707
-158.101.216.150 94.198.159.102 u 1101  1280  32.8930  35.2584 187.7465
-193.219.28.60   194.29.130.252   2 u 1090  1280  45.6136   5.0617 134.6725
-93.105.88.82194.146.251.101  2 u 1095  1280  60.3874  34.2942 141.3178
*91.227.212.78   194.146.251.100  2 u 1101  1280  60.9678  39.8529  37.1395

and no more than eight remain after reconnecting.

[tdu:debiantutor] ~
 date; ntpq -pn | grep -Fv pool
Sun  6 Aug 21:35:39 CEST 2023
 remote   refid  st t when poll reach   delay   offset   jitter
===
*78.47.168.188   17.253.14.1252 u4  128  377  42.4310  -4.7689  11.4233
+144.76.0.164193.190.230.65   2 u3  128  377  35.6690   0.3855  71.0345
+162.159.200.123 10.216.8.4   3 u   21  128  377  36.2527  -5.7169  19.8860
+86.90.148.127   125.199.53.762 u   14  128  375  48.7406  -0.6191  19.9919
+158.101.216.150 94.198.159.102 u  153  128  376  36.3177  -2.0115  23.1690
+193.219.28.60   210.100.177.101  2 u   15  128  377  47.5177   1.2094 134.4656
+93.105.88.82194.146.251.101  2 u   21  128  377  62.4123  -8.6088  18.5243
+91.227.212.78   194.146.251.100  2 u   15  128  377  59.2438   0.2789 182.5268


Something is undocumented here, if not wrong.

Feel free to tell me how I can assist in tracking the issue down.

To be clear, I am happy the number of servers is no longer 18, but confusion
does remain as to how exactly the "pool" directive works.

Best regards


-- 
Thure Dührsen
Kiel, Germany


Thure Dührsen
Elisabethstraße 116
4. OG rechts
24143 Kiel

Tel. +49 431 591 9707
Handy +49 160 9641 8332
tduehr...@schreib-doch-mal-wieder.de
ICQ 192-268-058
öffentlicher GnuPG-Schlüssel siehe http://www.informatik.uni-kiel.de/~tdu/