Conversation
Edited 1 year ago
According to my back-of-napkin calculations (looking at the number of archived messages in the past month, averaging it per day, and multiplying by number of subscribers):

- vger.kernel.org delivers about 4.5 million messages per day
- 70% of that is linux-kernel@vger, with about 3.2 million messages delivered per day
- a very remote second is netdev@vger (300,000 messages delivered per day)
- even a more remote third is kvm@vger (133,000 messages delivered per day)

Current migration stats:

Out of the total of 203 public vger lists:

- 48 lists will be sunset (due to obsolescence/inactivity)
- 40 lists are already migrated to new infra
- 115 still reside on legacy infra

The 40 migrated lists are about 788,000 messages daily, or ~20% of all traffic, so 80% of all mail traffic is still going through the legacy infra.

I hope to complete all migrations by this holiday season.
1
5
17

Thorsten Leemhuis (acct. 1/4)

Edited 1 year ago

@monsieuricon

sometimes I wonder if we still need LKML now that we have lore.kernel.org/all/

* LKML is CCed on a lot of stuff, but a lot of stuff it's not, so it's not a "everything" list/archive

* because so much is CCed on LKML it's impossible to read for nearly everyone, which makes it hard to see/follow stuff that has no list/no better place to post.

Or am I missing something here?

2
0
1

@jmorris @monsieuricon

afaik @corbet still tries to skim it

0
0
2

@kernellogger @monsieuricon A lot of things don’t have a more specific list to use so need just somewhere to post - it’s not filling a hugely different role to patches@lists but it is there.

1
0
0

@broonie @monsieuricon

yeah, that's why I wrote "hard to see/follow stuff that has no list/no better place to post"

So maybe the real best solution to the situation it to stop CCing LKML on stuff where a suitable list exists

0
0
1