Messages

forgot to mention, please consider: boot time in my measurings is NOT hardware boot, but up from the OPNS/BSD bootloader timeout. my OPNS is always hanging on "Configuring firewal..." for 5-6 minutes, then goes again with the interaface list, then hanging again on "configuring firewall..." for another 2-3 minutes.

Hello,
since ever i had a very long boot time, speaking about 8-10 minutes!
always i thought it was the hardware or some limitations from virtualization. Since i use mostly commodity HW and ProxmoxVE, i thought it was an issue with old Xeon 2690v3 (2.6GHz base / 3.50 GHz turbo) because of low frequency... now OPNS is running on a dual Epyc 7252 (3.1GHz base / 3.2GHz turbo) and has 256GB of DDR4 RAM physically (to mitigate the slow reboot i tested how it is going on a supemicro server), it is still booting slowly as og 8-10mintes!
In my case OPNS is only managing 73 VLANs and 85 IPsec tunnels.
nothing changed actually. so i assume it was not the PVE virtualization and not the old Xeon cpu.
Asking for ideas and advices, where to look. it is not about the network speed, it is about the boot time. thanks in advance!

reset all local conflicts did it. tried only plugin conflicts. thanks you!

Hello,
I bought a new DEC850, which was preinstalled with Business Edition, since this is for home use, i decided to go with community edition, so changed repository type, updated to the latest.
Now there is a plugin missing os-OPNBEcore, which obviously belongs to the Business Edition, how can i get rid of it, or may i just ignore it?
Thanks in advance

Sensei will stop working when detected that a partition is %90 + full. But did you look that which folder is growing fastly? It is not expected behavior for DB data.

thank you. it is indeed mongodb index files (/usr/local/datastore/mongodb/*.wt, a lot of them), i also found something similar in the mongodb community https://www.mongodb.com/community/forums/t/mongodb-disk-space-increases-abruptly-wiredtigerlas-wt/3620
will investigate further.
also i am now a little more relaxed by the 90% disk limit, in case i am not around, when it grows again.

Hello,
i am evaluating Sensei for some weeks. The appliance has a disk of 64GB, 20% used. With Sensei + MongoDB 23% is used (when erase all reporting data), max number of days is 1.
Now the interesting part: disk usage grows slowly (3-4 days) to about 50%, then all of a sudden 70-75% and growing very fast to 80%++ (within 1-2 hours), where i overreact and reset reporting data again. After erasing, back to 23% disk used.
My question: what happens, when i would not reset the reporting data in Sensei? Is there an automatic cap of used disk space before the disk would run to 100% full? would anything crash? i had no nerves to test that in the production environment. my testing environment has not enough traffic to get into such scenario.
Thanks in advance

edit:
all newest stable updates installed (opns 21.1.8, sensei enigne 1.9.2, ui 21.7.7, db 1.9.21070514)
issue persist since i installed sensei 8-9 weeks ago
sensei is in passive mode, level warning, deployment size small, rotation 1 day, max days to store reporting data 1 day (not a lot of traffic except the usual attacks from outside)

Lieber Frank,

vielen Dank für das os-qemu-guest-agent Plugin. Mir sind die fehlenden Features in Eugens (Aborches) FreeBSD Port bekannt.

Женя, в первую очередь тебе огромнейшее спасибо за портирование и что добавил в репозиторий FreeBSD Ports! ПС: второй лог добавил для референции, думаю тут большинству ясно, что fs команды отсутствуют.

Ich habe das Plugin gestern installiert (OPNS 21.1.6) und es funktioniert auch soweit sehr gut! Nur ein Problem, dass es beim OPNS Boot nicht startet (Logs anbei). Allerdings, sobald die OPNS online ist (auch 20min abgewartet nach dem Boot), unter Services - QEMU Guest Agent - Settings -> Klick auf "Apply" ist sofort online und in Proxmox GUI die IPs sichtbar (sonst steht da das bekannte: "guest agent not running").
Evtl. gibt es eine Möglichkeit den GA später zu starten, dass die Abhängigkeiten erfüllt sind, falls es daran liegt. Logs...
1. ohne Debug Login, sofort nach dem Boot (und nach Klick auf apply kommt nicht mehr, aber GA funzt dann)
2. mit aktiviertem Debug Logging (ist zu erwarten, dass die fs-features sich melden, sonst aber nichts neues mehr drin)

Danke für den Link. Laut YT ist er erst vor 3 Std. geuppt :)
Es lebe OPNSense!

I would like to add, with virtualized OPNSense 19.7.3 i have the same issue (since 17.x).
Hardware:...
Proxmox 5.1.x - 6.0.6, MultiWAN, Virtio and Intel E1000 virtual NICs tested (not tested with virtualized Realtek or Vmxnet3)
The physical nodes tested HP DL580g7, DL380g6/g7/g8, DL360g8/g9
Physical NICs: onboard (whatever those nodes have: broadcom, netxen, intel) + different intel pci-e cards (currently nodes are running with intel x520, tested with x540 and some other cards, also with some silly cards like chelsio, which have different issues at all).
as soon as PPPoE reloads -> crash (reproduce: interfaces / overview / wan / reload OR disconnect)
since my ISP is not forcing regularly reconnects, its not that critical, but still painful :)

Edit: changed the ISP to one w/o pppoe.