Replies: 5 comments
-
We run with a few more IRRs sources than you have listed, and with RPKI enabled on most. |
Beta Was this translation helpful? Give feedback.
-
I'm running IRRd 4.2.5 with 4 cores, 32 GB RAM, 512 GB Samsung 840 SSD with these databases:
|
Beta Was this translation helpful? Give feedback.
-
😭😭😭
Where is TC?
😭😭😭
Em dom., 24 de jul. de 2022 09:22, Job Snijders ***@***.***>
escreveu:
… I'm running IRRd 4.2.5 with 4 cores, 32 GB RAM, 512 GB Samsung 840 SSD
with these databases:
irr$ echo '!j-*' | nc irr.yycix.ca 43
A264
RPKI:N:-
ARIN:Y:0-358693
RIPE:Y:0-53859417
APNIC:Y:0-10232537
LACNIC:Y:0-422200
AFRINIC:Y:0-1216604
NTTCOM:Y:0-1297659
RADB:Y:0-13644487
CANARIE:Y:0-4470
LEVEL3:Y:0-835712
ALTDB:Y:0-95297
JPIRR:Y:0-222710
IDNIC:Y:0-199263
BELL:Y:0-249600
RIPE-NONAUTH:Y:0-53859417
—
Reply to this email directly, view it on GitHub
<#669 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AB46SYLBQOVH66RZXBTJZNLVVUYQDANCNFSM54KZ2RPA>
.
You are receiving this because you are subscribed to this thread.Message
ID: ***@***.***>
|
Beta Was this translation helpful? Give feedback.
-
I run a main branch instance with NTTCOM,RADB,RIPE,RIPE-NONAUTH,ALTDB,BELL,LEVEL3,RGNET,APNIC,JPIRR,ARIN,BBOI,TC,AFRINIC,RPKI,PANIX,OPENFACE,NESTEGG,CANARIE,LACNIC with RPKI enabled for all, on this quad core celeron, 32GB ram, some scavenged SSD. I also use it for a few other things. My /var/lib/postgres is 30GB. I also have a 4.2 and 4.1 instance on 8GB each. I have 5 HTTP and 5 whois workers running (each of these needs ~200MB). Initial imports also need more memory, so you could probably limit peak use by not adding all the large ones at the same time. On the 8GB nodes this is something I have to do if I need to run a full import of multiple sources. |
Beta Was this translation helpful? Give feedback.
-
Thank's all for the replies. Based on your feedback I see that disk space and RAM are the most important resources for the IRRdv4 daemon. @mxsasha indeed when I bootstrap IRRd (with recommended resources) it consumes significantly more resources and I need to constantly feed the VM with memory and disk. However, after a week it stabilizes and resource consumption remains at the same levels. |
Beta Was this translation helpful? Give feedback.
-
Hi Sasha/all,
I was wondering how much resources do I need to provide to my VM if I want to mirror all major IRRs (RIPE,ARIN,LACNIC,APNIC,AFRINIC) and have IRRd running stable?
I believe that the recommended resources listed in the documentation are not enough. Especially the RAM size and the disk space required for the postgress.
Did any member of the community have a successful deployment with all major IRRs that can share the specs of his node?
Kind Regards
Stavros
Beta Was this translation helpful? Give feedback.
All reactions