Hi Daniel & Michael,

Many thanks for replying.

On Tue, Dec 9, 2025 at 9:18 AM Daniel P. Berrangé <[email protected]> wrote:
>
> On Tue, Dec 09, 2025 at 02:24:44AM -0500, Michael S. Tsirkin wrote:
> > Hi guys,
> > who has ability to contact the list admins? Can help Salil?
>
> I'm not a list admin, but I have some observations below
>
> > ----- Forwarded message from Salil Mehta <[email protected]> -----
> >
> > From: Salil Mehta <[email protected]>
> > Date: Tue, 9 Dec 2025 00:44:04 +0000
> > To: "Michael S. Tsirkin" <[email protected]>
> > Subject: [ATTENTION] Mails on qemu mailing list are disapapearing making 
> > communication unreliable
> > Message-ID: 
> > <caj7pxey8dpk53epe6ycxqhpmxd9b7e9ty49ij6hzgu450og...@mail.gmail.com>
> >
> > Hi Michael,
> >
> > I hope you are doing well.
> >
> > I’ve noticed that some emails are going missing from the qemu-devel mailing
> > list. This has now happened multiple times, both with emails I have sent 
> > from my
> > official address and with emails from other contributors.
> >
> > Over the past three months, I have consistently observed emails disappearing
> > from the qemu-devel archives, which makes the mailing list communication 
> > highly
> > unreliable.
>
> I can see patches arriving in my INBOX via qemu-devel from 
> [email protected]
> but no patches from [email protected]
>
> I do see some (non-patch) emails from [email protected], but the
> mailing list is re-writing this From address for DMARC countermeasures
>
>  Salil Mehta via <[email protected]>
>
>
> Recapping mail delivery policies, DMARC requires a pass for *either*
> SPF or DKIM checks.
>
> SPF checks will always fail for mails forwarded via any mailing list.
>
> DKIM checks should still succeed *provided* the mailing list does not
> modify the message content, or certain headers. QEMU mailing list
> behaves in this respect generally, and applies countermeasures when
> appropriate.
>
> I see no  DKIM records huawei.com, but I do see SPF records.


Thanks for identifying this.

When you say there are no DKIM records for huawei.com, do you mean
that my @huawei.com mails do not contain any DKIM-Signature header,
or that a DKIM-Signature is present but cannot be validated against
a published key?

If there are SPF/DKIM/DMARC logs on the list side, would it be
possible to see whether DKIM validation is failing for my messages?


>
> This guarantees DMARC fail for any @huawei.com sender using any mailing
> list, however, the @huawei.com DMARC policy is only set to 'quarantine'.

Agreed. I will check with the IT department here about the DKIM
headers and whether the public key is being published correctly.


>
> That should not block delivery to list subscribers, merely suggest
> that MTAs put the mails in the Spam/Junk folder *if* DMARC fails.
>
> mailam is cautious and thus applying DMARC countermeasures to huawei.com
> addresses due to he missing DKIM records, in order to reduce the chance
> that the mails end up in Spam/Junk folders for list subscribers.
>
>
> This does not explain why messages might never appear in the list to
> begin with though.

Agreed.

>
> It is notable, however, that messages from [email protected]
> do arrive if they are non-patch mails. Presumably those are via a
> normal email client

This is not always true.


>
> I wonder how Salil is sending patches ? Some tool like git send-email
> or git-publish ?

Mails disappear or not appear in the mailing list even when I reply
via outlook and the same happens when I sometimes use git-send-mail.

check these:
https://lore.kernel.org/qemu-devel/[email protected]/

AND

https://lore.kernel.org/qemu-devel/caj7pxea721f4ewqpddcztgron-rsb2tqthf-_kb+mmsrkk1...@mail.gmail.com/

I'm replying to this broken link which existed but disappeared later.
 [not found] <[email protected]>


>
> If so I then wonder if the tool is NOT correctly configured to use an
> official huawei.com outbound SMTP relay ?
>
> While you can have git send-email deliver directly to lists.nongnu.org,
> the huawei.com SPF policy will almost certainly result in the mails
> getting either rejected or dropped, *UNLESS* they are relayed via one
> of the officially designated huawei.com SMTP relays.

This configuration is correct. With this configuration, no mail will be sent
outside Huawei. There are other addresses as well in the TO/CC list.
I'll rule out this aspect.


> > For example, the attached .eml file is a message from Vishnu (Ampere) which
> > was sent to the list but later disappeared. This is starting to raise 
> > serious
> > concerns about the reliability of the mailing list, and the broken 
> > communication
> > is creating confusion for everyone involved.
>
>
> In terms of mails from Vishnu, I see nothing in my INBOX from Vishnu.
>
> The mail quoted below has a From address of [email protected],
> while people seem to sending to/cc for [email protected]
>
> Is that From address correct ? It seems like the latter is more likely to be
> the canonical address for Ampere accounts.

It must be the correct one because I attached the .eml for the received mail.
And I really don't have a clue how this address is different from  the
other i.e.
@os.amperecomputing.com

[+] Adding Vishnu to CC, maybe he can share more details and share his
.eml file.


>
> If not, perhaps that is triggering some rule that discards the mails when
> sending is amperemail.onmicrosoft.com

Could be. Do we have access to the logs to verify the reason?


Best regards
Salil.

> > From: Vishnu Pajjuri <[email protected]>
> > Date: Mon, 27 Oct 2025 13:58:59 +0530
> > To: Igor Mammedov <[email protected]>, [email protected]
> > Cc: [email protected], [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected],
> >  [email protected], [email protected],
> >  [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected], [email protected],
> >  [email protected], [email protected]
> > Subject: Re: [PATCH RFC V6 00/24] Support of Virtual CPU Hotplug-like 
> > Feature
> >  for ARMv8+ Arch
> > Message-ID: 
> > <[email protected]>
> > In-Reply-To: <20251006160027.20067fe4@fedora>
>
>
> With regards,
> Daniel
> --
> |: https://berrange.com      -o-    https://www.flickr.com/photos/dberrange :|
> |: https://libvirt.org         -o-            https://fstop138.berrange.com :|
> |: https://entangle-photo.org    -o-    https://www.instagram.com/dberrange :|
>

Reply via email to