Admin

Admin

admin@libera.site

This channel has not added a profile description yet

2026-01-08 10:41:22

Admin

admin@libera.site

@Mark Nowiasz
Hi!
could you check what is going on your stats?
My scripts read weird data(and the problem is only on your instance):
20260101:1767248406,hub.netzgemeinde.eu,2708,508,186,0,257813,0,1,false
20260102:1767334803,hub.netzgemeinde.eu,2709,505,186,0,258486,0,1,false
20260103:1767421203,hub.netzgemeinde.eu,2718,511,188,288472,259349,0,1,false
20260104:1767507604,hub.netzgemeinde.eu,2718,511,181,288779,260290,0,1,false
20260105:1767594004,hub.netzgemeinde.eu,2720,512,182,0,261257,0,1,false
20260106:1767680403,hub.netzgemeinde.eu,2722,514,184,289516,262255,0,1,false
20260107:1767766803,hub.netzgemeinde.eu,2724,514,186,0,0,0,1,false
20260108:1767853203,hub.netzgemeinde.eu,2729,516,190,290397,264443,0,1,false

The problem is 6th row, statuses. Sometimes zero, sometimes nearly 300k. 🤷‍♂️
Then it makes weird graphs:
#^https://libera.site/item/af7db1ec-3c1d-454e-a7dd-a5892fa21736

2026-01-08 11:00:28

Mark Nowiasz

buckaroo@hub.netzgemeinde.eu

@Admin No idea. Well, occasionally I got alerts that php-fpm/hubzilla is down. Either there's a huge load at the time (but soon to resolved) or all fpm workers are in use. There are a couple of explanations of this: A remote site is slow therefore slowing down the fpm processes (when cloudflare was down it affected the site too, for this very reason), therefore the maximum limit of fpm process has been reached, so there's a gateway timeout for new processess and so on. Since this is only occasionally it's next to impossible to debug - and nothing I can do about it, there's only a limit amount of memory/cpus, so raising the number of fpm workers will only result in extremely slow performance before reaching the limit, too

2025-11-06 20:40:18

Admin

admin@libera.site

How to set higher limit of notifications in widget:
util/pconfig 3 system notifications_count_limit 1000

3 is channel ID.

Thanx @Mario Vavti 💪