Re: [mailop] warming up IPs, Microsoft?

2023-03-06 Thread Benny Pedersen via mailop
John R Levine via mailop skrev den 2023-03-06 18:55: Linode has a bunch of different IP address blocks and I would expect recipients to block the ones that send annoying amounts of spam. That's what I do. So as likely as not, you're just lucky that you don't have annoying neighbors. linode do

Re: [mailop] warming up IPs, Microsoft?

2023-03-06 Thread John Levine via mailop
It appears that Laura Atkins via mailop said: >In the B2C space domain reputation is more important than IP reputation >anyway. > >You may also want to look at SSDNodes for VPSes. Looks fantastic except for the IPv6-only part. I see the larger ones have both v4/v6, makes sense if you want ser

Re: [mailop] warming up IPs, Microsoft?

2023-03-06 Thread Michael Rathbun via mailop
On Mon, 6 Mar 2023 10:52:35 +, Laura Atkins via mailop wrote: >I have had a number of clients over the last 3 or 4 years using SES without >any delivery problems that we could attribute to the IP addresses. Once we ran >through fixing the things under their control, delivery was great. I

Re: [mailop] warming up IPs, Microsoft?

2023-03-06 Thread John R Levine via mailop
Huh. We don't have any issues sending email to them from Linode, including a small number from one of our new IP addresses I've been trying to warm up. Linode has a bunch of different IP address blocks and I would expect recipients to block the ones that send annoying amounts of spam. That's

Re: [mailop] warming up IPs, Microsoft?

2023-03-06 Thread Mark Fletcher via mailop
On Mon, Mar 6, 2023 at 6:36 AM John Stoffel wrote: > > I've been using digital ocean and then moved to Linode to home my > personal domain @stoffel.org, but then I'm spending time looking to > move again because charter.net is my town Cable company and a bunch of > neighbors use @charter.net emai

Re: [mailop] warming up IPs, Microsoft?

2023-03-06 Thread Atro Tossavainen via mailop
> I believe it, but the more relevant question is what fraction that is of the > total > mail they send. I see way more real mail than spam from them. I can only speak to the mail we see. I am sure all of the entities that are sending to our spamtraps mostly send good email. I simply could not h

Re: [mailop] warming up IPs, Microsoft?

2023-03-06 Thread Laura Atkins via mailop
> On 5 Mar 2023, at 21:53, Mark Fletcher via mailop wrote: > > On Sun, Mar 5, 2023 at 1:15 PM John R Levine > wrote: >> >> If you need a big VM there's always AWS. They do a surprisingly good job >> of managing outbound mail. You get 62K messages/mo for free, then 1

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread John Levine via mailop
It appears that Atro Tossavainen via mailop said: >In February we had close to 6000 separate domains sending from >AWS SES to our spamtraps. I believe it, but the more relevant question is what fraction that is of the total mail they send. I see way more real mail than spam from them. They're

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread Mark Fletcher via mailop
On Sun, Mar 5, 2023 at 1:15 PM John R Levine wrote: > > If you need a big VM there's always AWS. They do a surprisingly good job > of managing outbound mail. You get 62K messages/mo for free, then 10c per > 1000 messages sent from a VM. > For the amount of email we send, that cost structure wo

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread Atro Tossavainen via mailop
> You have to validate each domain you use for sending, which is a > modest pain, but that's one of the reasons their mail stream is > pretty clean. Do you mean AWS SES specifically? They're consistently #4 by volume in our dataset (occasionally even #3, rarely #5), next only to SendGrid, ExactTar

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread John R Levine via mailop
It also occurs to me that you don't need to do your computing and mail on the same VM. Mail is rather lightweight so you could run a mail server at Tektonic, and send messages from other places via port 587 submission. On Sun, 5 Mar 2023, Mark Fletcher wrote: On Sun, Mar 5, 2023 at 10:20 AM

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread John R Levine via mailop
Thanks for the recommendation; unfortunately they wouldn't work for us. Their largest VM is less than half the size we would need for our databases, also they don't appear to have an API to provision new VMs. If you need a big VM there's always AWS. They do a surprisingly good job of managing

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread Mark Fletcher via mailop
On Sun, Mar 5, 2023 at 10:20 AM John R Levine wrote: > > I've been happy with a small provider called Tektonic. If you've never > heard of them, that's a good sign. > > Thanks for the recommendation; unfortunately they wouldn't work for us. Their largest VM is less than half the size we would ne

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread John R Levine via mailop
On Sun, 5 Mar 2023, Mark Fletcher wrote: Best I can tell, in our 9+ years, being hosted by Linode has never been an issue wrt deliverability, and as a hosting provider, they've been nothing but responsive and reliable. That said, they were recently bought by Akamai, and have just raised prices.

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread Bill Cole via mailop
On 2023-03-05 at 11:47:07 UTC-0500 (5 Mar 2023 11:47:07 -0500) John Levine via mailop is rumored to have said: I wouldn't try to send mail from Linode if I cared about it being delivered. Indeed. Linode is a standout amongst cheap VM providers in the supply of attack traffic they make avail

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread Mark Fletcher via mailop
On Sun, Mar 5, 2023 at 8:53 AM John Levine via mailop wrote: > > I am reasonably sure that no serious mail system uses UCEPROTECT as > anything > other than a very weak signal. It's not surprising they list what other > people list, since they list about half the net. > > Agreed. I think one of

Re: [mailop] warming up IPs, Microsoft?

2023-03-05 Thread John Levine via mailop
According to Simon Greenwood via mailop : >All those IPs are reported by UCEPROTECT and Polspam at the moment, and >I have seen that the 800lb gorillas seem to refer to UCEPROTECT, or at >least their findings coincide. I am reasonably sure that no serious mail system uses UCEPROTECT as anything