Re: Emails to multiple recipients on same server not getting delivered

2021-04-06 Thread Linda A. Walsh

On 2021/04/05 12:27, John Stoffel wrote:

I suspect you're only delivering to the first recipient, in the email,
and not calling the delivery for each and every local recipient
individually.
  

--
I'll 2nd that, as from what you've said, you took the mailer
that handles distribution of each copy of the email to
all and multiple recipients out of the equation.





Re: Emails to multiple recipients on same server not getting delivered

2021-04-05 Thread Linda A. Walsh

On 2021/04/05 11:24, John Stoffel wrote:

"Steve" == Steve Dondley  writes:



Steve> When I send an email to a single user on a server, it is received by the 
Steve> user without a problem. But when sending to multiple users, the emails 
Steve> disappear into a black hole. The logs contain no errors and indicate the 
Steve> emails were sent:
  

-
   FWIW -- if one or more of those users have their email accounts hosted
at google, they have a practice of deleting multiple copies of an
email that would goto different addresses.  Since all of those emails
are the same, they wanted to ensure they only needed to store it once.

   (this is similar to the problem of people sending to the original author
and to a list.  Thanks to the google change, usually the 2nd copy that
takes a bit longer due to going through list software) This disregards
instructions to send dups of the message to other addresses.




Re: Btrfs RAID-10 performance

2020-09-15 Thread Linda A. Walsh

On 2020/09/10 07:40, Miloslav Hůla wrote:
I cannot verify it, but I think that even JBOD is propagated as a 
virtual device. If you create JBOD from 3 different disks, low level 
parameters may differ.
  


   JBOD allows each disk to be seen by the OS, as is.  You wouldn't
create JBOD disk from 3 different disks -- JBOD would give you 3 separate
JBOD disks for the 3 separate disks.

   So for your 16  disks, you are using 1 long RAID0?  You realize
1 disk goes out, the entire array needs to be reconstructed.  Also
all of your spindles can be tied up by long read/writes -- optimal speed
would come from a read 16 stripes wide spread over the 16 disks.

   What would be better, IMO, is going with a RAID-10 like your subject
says, using 8-pairs of mirrors and strip those.  Set your stripe unit
for 64K to allow the disks to operate independently.  You don't want
a long 16-disk stripe, as that's far from optimal for your mailbox load.
What you want is the ability to have multiple I/O ops going at the same
time -- independently.  I think as it stands now, you are far more likely
to get contention as different mailboxes are accessed with contention
happening within the span, vs. letting each 2 disk mirror potentially doing
a different task -- which would likely have the effect of raising your
I/O ops/s. 


   Running raid10 on top of raid0 seems really wasteful




Re: Cannot delete folder

2018-05-21 Thread Linda A. Walsh

Yves Goergen wrote:

To be honest, I can't follow you. I see that I have inconsistent
permissions (but all the same owner and group). I did use a different
mail server software in the past, but that's long ago and I can't even
remember when I switched to Dovecot. I think it was Courier-IMAP before.

Maildirs for new mailboxes are still created by one of my own scripts so
maybe that script creates the maildirs with some wrong permissions. I'll
first have to look up what all those funny letters actually mean.
They're too rare to remember.

---
hopefully quick explanation of letters 
  +-d=directory

  | usr(me) (r=read,w=write,x=cross(ing)...allows
   seeing things below this directory
   but 'r' controls seeing things *in*
   this directory)
  | ||| group
  | ||| ||| everyone else
My base permissions were : d rwx rwx r-x
most were: :   d rwx rws r-x  The 's' on the group-triplet means
 set-group on files+dirs below this

The exact permissions aren't so important other than to know that
1) when it creates an internal directory (like .imap/whatever), those
  it followed the permissions on my original mbox.
2) when it creats an internal mbox (like INBOX, Trash, Sent...etc) it may
   use whatever it's own umask allows.  Umasks are often set to system 
defaults
   like 022 or 002 on 'friendly systems -- they *remove* a bit from a 
permission set on a file.  Those perms are in octal (0-7), with octal 
you need
3 bits 0111 = 1*4 + 1*2 + 1*1 -- the high bit (multiplied by highest 
amount, in this case 4, corresponds to the read permission, the 2 
corresponds to write permission and the 1 corresponds to execute on 
files files, or cross-through
for directories.  So a umask of 022 would filter out the 2nd bit in each 
permission mask (i.e. the write bit for groups (the middle number) and 
the write
bit for 'other' (the 3rd group)  Since I allow groups to write, I have 
002 so
people in the same group can write. 


   Windows will use something like 077 on your home directory -- only the
user has any access, by default.  the 7's mask out access for any 
created files.


If the process that runs dovecot runs with a umask different from you, 
by default it might create differently permissioned files.  If you 
create a directory manually, that also might be different from what you 
normally see.


   In your case 'Trash' had different permissions -- a directory or 
file that would have been created by dovecot.  so it may be running with 
different default

permissions (settable via the usmask), than you.

   Hope that helps some rather than confuses more...was a quick summary 
of stuff.









Re: Cannot delete folder

2018-05-21 Thread Linda A. Walsh

Yves Goergen wrote:
The issue still exists. Can anybody explain to me why dovecot creates 
IMAP folders with the wrong filesystem permissions?
On a lark, I looked through my dirs @ permissions.  Shorted lines a bit 
so they'd fit w/o extra lines between them using:

(get rid of text before permissions, and shorten user/group to a few letter)
find . -type d -ls|sed -r 's/^\s*\S+\s+\S+\s+// ; s/linda(group)?/usr/g'

Made it wasy to look down the 1st column and find where things were 
different.

FWIW, using mbox format.

My base permissions were : drwxrwxr-x
most were:   : drwxrwsr-x
I noticed that subdirs I
created *by hand* (to later
move folders into), often
had different permissions:  drwxrwsr--

The first auto-dir created
by dovecot (.imap under
top dir) had some perms
drawn from the base  : drwxrwxr--
First auto-subdir created
(INBOX) had  : drwxrwxr-x
  BUT...had a different
group  -- one that I used
ages ago before making sure
that the username based group
took priority.
(I make each user's primary
group one based on their
username).

Another imap created subdir 
(.imap/Junk) : drwxrwx---


took it's perms from the base
file.  Most of my mbox files
have : -rw-rw-r--
but Junk was : -rw-rw (thus no 'r' bit on .imap/Junk)


In all the cases, the folders that were different
derived from manually created mboxes, *OR* a few
auto-created MBOXES (like INBOX ... and likely 'Trash' --
but before using dovecot, I already had a 'Trash' mbox, so it
just used it w/the same perms.

In your case, the permissions likely derive from the
umask that dovecot was(is?) running with when it created
the 'builtin' files (like Trash, Inbox, etc..)

Does that fit your case?







Re: Dovecot under Linux with mail-extension and case insesitiv

2015-08-06 Thread Linda A. Walsh

Sascha wrote:

Hello,

i use doevecot 2.2.18 current. My Problem is with email-extension and
case sensitiv folders.
  

I've had similar problems.  As I run Win-7 against a Samba-linux server.
Win-7 does case-preserving+case-ignore -- samba does case Mangling -- it 
seems
to uppercase machine names  lowercase user names... causes alot of 
chaos


I'm only 1-user, so I changed my /etc/passwd file to have 2-3 names for
each   user-id I wanted to show up:
root:x:0:0:root:/root:/bin/bash
Bliss\root:x:0:0:root:/root:/bin/bash
BLISS\root:x:0:0:root:/root:/bin/bash

At times/over the years, I've had Samba give me each variation at least
once. But it's not just samba... different brands of 'ssh' will pass
different things -- some just 'user', some Domain\User, some 
Domain/User(cygwin).  Since most of my personal users are all lower

case, they don't usually need special attention, but in home directories
under 'Bliss', have a symlink BLISS-Bliss.  On some shares also needed
another symlink bliss-Bliss.  (ARR)...(BTW - Bliss was meant as
irony... so symlinks might help but will look ugly)...it can get
worse -- in user/group names, sometimes spaces are removed..
'Power Users' = powerusers.. ARG!!!

But when I've talked to dovecot... I always have just needed the
user name (no Domain)... Maybe your email software has some setting
to tell it to play nice?

Sorry for the bad idea for large-scale implementation, but it was
easiest to do for a few users..(home network)...

BTW... some file systems are moving toward ignoring case...
xfs has had a case-ignore option but only works in ascii range.
Heard apple' implemented case-preserve+ignore on theirs,
and seem to remember reading somthing about Solaris(Sun/Oracle)'
ZFS getting a case-ignore+preserve option.

Good luck!



Example:
user+extenstion will be delivered to the user and subfolder extenstion
so this is okay.
but user+extenstion will not be delivert to the user and exiting
subfolder Extenstion so that is not okay.
The mail will be also delivered in the subfolder extension. The exiting
Folder Extension will not used.

Give's a way to tell dovecot, use the extension as a Folder, if a
exiting folder with Upper/Lower Cases will be detect and use the right
folder to delivery?

 -- Regards Sascha

---
Diese E-Mail wurde von Avast Antivirus-Software auf Viren geprüft.
https://www.avast.com/antivirus