isync-devel — All-purpose list, actually - but RTFM first. Moderated for non-subscribers for SPAM protection.

Re: mbsync not obtaining recent emails

[Amulya M. <amu...@fa...>]

I am using aerc for viewing the emails, and the emails display -- there
is simply a large gap in what emails were being fetched. I have attached
the mbsync -D log and the isyncrc.

Re: mbsync not obtaining recent emails

[Oswald B. <osw...@gm...>]

On Thu, Mar 13, 2025 at 11:11:23PM -0400, Amulya Mohan wrote:
>How do I get the emails from between March 04 and 1 hour ago?
>
have you tried an interactive imap client (say, thunderbird) to see
whether the mails are exposed via email at all? if not, then there is
nothing isync can do about it.
if the mails are visible, then run mbsync with -D and mail me the log
along with the config file.

Re: mbsync not obtaining recent emails

[Amulya M. <amu...@fa...>]

Ok, so now mbsync is fetching recent emails, but the emails it fetched                                                                   
are from 1 hour ago. so... the emails in my inbox are:                                                                                   
                                                                                                                                         
[1 hour ago]                                                                                                                             
...                                                                                                                                      
[1 hour ago]                                                                                                                             
                                                                                                                                         
[March 04]                                                                                                                               
                                                                                                                                         
How do I get the emails from between March 04 and 1 hour ago?     

mbsync not obtaining recent emails

[Amulya M. <amu...@fa...>]

Hi,

I'm trying to setup mbsync for my fastmail account. I successfully created a channel and a Near and Far store, and i successfully ran `mbsync -a`. However, the emails that got pulled were an archived copy of my inbox from March 04, and the emails are wildly out of date. How do I get mbsync to pull the most recent emails from my folders?
help me, i googled around and found nothing

Amulya

[commit] tag 'v0.9' created

[ossi <os...@us...>]

The tag 'v0.9' has been created at d2b39d2.

[commit] tag 'v0.9.1' created

[ossi <os...@us...>]

The tag 'v0.9.1' has been created at 4390613.

[commit] tag 'v1.5.1' created

[ossi <os...@us...>]

The tag 'v1.5.1' has been created at ffc0884.

[commit] tag 'v0.9.2' created

[ossi <os...@us...>]

The tag 'v0.9.2' has been created at 5a44a33.

[commit] master: release preparation

[ossi <os...@us...>]

commit ffc0884efe4b688901c17270c1c0acca2d5d490e
Author: Oswald Buddenhagen <os...@us...>
Date:   Tue Mar 11 19:47:48 2025 +0100

    release preparation
    
    update NEWS and AUTHORS

 AUTHORS |  3 +++
 NEWS    | 23 +++++++++++++++++++++++
 2 files changed, 26 insertions(+)

diff --git a/AUTHORS b/AUTHORS
index 024f512..2323bc8 100644
--- a/AUTHORS
+++ b/AUTHORS
@@ -71,6 +71,7 @@ Honorary Contributors
 Alessandro Ghedini <gh...@de...>
 Andreas Grapentin <an...@gr...>
 Aurélien Francillon <aur...@eu...>
+Behnam Lal <dev@behnamlal.xyz>
 Ben Kibbey <bj...@lu...>
 Caspar Schutijser <ca...@sc...>
 Cedric Ware <ced...@no...>
@@ -87,12 +88,14 @@ Jaroslav Suchanek <jar...@gm...>
 Jeremie Courreges-Anglas <jc...@op...>
 Klemens Nanni <kn...@op...>
 Lorenzo Martignoni <lor...@te...>
+Ludovico Gerardi <lud...@po...>
 Magnus Jonsson <bi...@ac...>
 Marcin Niestroj <mac...@gm...>
 Martin Stenberg <ma...@gn...>
 Mike Delaney <md...@lu...>
 Nicolas Boullis <nbo...@de...>
 Nihal Jere <nihal@nihaljere.xyz>
+Paymon MARANDI <dar...@gm...>
 Reimar Döffinger <Rei...@gm...>
 Remko Tronçon <re...@el...>
 sb...@us...
diff --git a/NEWS b/NEWS
index 08c3049..22f3ab1 100644
--- a/NEWS
+++ b/NEWS
@@ -1,3 +1,26 @@
+1.5.1 (2025-03-11)
+==================
+
+Improvements:
+
+- mbsync-get-cert now supports STARTTLS; new option -s
+- Zero-sized messages from IMAP Stores are now accepted
+- UIDVALIDITY change recovery is now attempted even if both sides of
+  the Channel are affected
+- The sync summary at the end is more concise again
+- Cosmetic improvements to some console output
+
+Bug Fixes:
+
+- Fixed IMAP INBOX not being properly recognized with some servers
+- Fixed Maildir INBOX nested into Path not being implicitly listed
+- Fixed crash when resuming message propagation with MaxMessages
+- Fixed --list-stores hanging after synchronous error
+- Fixed --dry-run without --debug-driver not being really dry
+- Fixed building from pristine git clones
+- Fixed building from shallow git clones
+
+
 1.5.0 (2024-08-02)
 ==================
 

[commit] master: expand NEWS

[ossi <os...@us...>]

commit ea22b09ac77380c6f530c09a5dc52dc2810f83c3
Author: Oswald Buddenhagen <os...@us...>
Date:   Tue Mar 11 19:16:59 2025 +0100

    expand NEWS
    
    include bug fixes and other improvements, as otherwise they wouldn't be
    documented in a user-accessible way, given that ChangeLog wasn't all
    that usable, and is gone entirely now.
    
    this backfills from the release notes i have put next to each tar ball
    since 1.0.5. and then some edits, in particular release dates.

 NEWS | 690 +++++++++++++++++++++++++++++++++++++++++++++++------------
 1 file changed, 556 insertions(+), 134 deletions(-)

diff --git a/NEWS b/NEWS
index 0438992..08c3049 100644
--- a/NEWS
+++ b/NEWS
@@ -1,229 +1,651 @@
-[1.5.0]
+1.5.0 (2024-08-02)
+==================
+
+Compatibility Concerns:
+
+- The reference point for relative local paths in the configuration file
+  is now the file's containing directory
+- Maildir Path cannot be nested into Inbox anymore (this was never documented)
+- Renamed the ReNew/--renew/-N options to Upgrade/--upgrade/-u and
+  Delete/--delete/-d to Gone/--gone/-g
+- Superseded SSLVersions option with TLSVersions, and disabled TLS v1.0
+  and v1.1 by default
+- Renamed SSLType option to TLSType
+- Placeholders will be now created for messages exceeding MaxSize even if they
+  are flagged on the source side
+- Placeholder upgrades no longer pull flag updates along unless also requested
+- New messages which we are about to expunge from the source side are not
+  propaged any more even if the target side would keep them
+- Tunnel is now consistently assumed to be secure, so some warnings are gone
+
+New Features:
+
+- Changed default config & state locations to follow the XDG basedir spec;
+  the old locations remain supported
+- Added support for IMAP mailbox names with non-ASCII characters
+- Added support for Maildir Paths with suffixes (not ending with a slash)
+- Made the Channel side to expire with MaxMessages configurable
+- MaxMessages and MaxSize can be used together now
+- The unfiltered list of mailboxes in each Store can be printed now
+- A proper summary is now printed prior to exiting.
+  This includes expunges, which are now included in the progress as well.
+- Added support for mirroring deletions more accurately; option ExpungeSolo
+- Added new sync operation 'Old' to retry previously skipped messages
+- Added --ext-exit option to indicate with the exit code whether Stores
+  were modified
+- Added --dry-run option
+
+Improvements:
+
+- Added support for the LITERAL- IMAP extension, which improves upload
+  performance with f.ex. GMail somewhat
+- Improved error handling when attempting to store too big messages on
+  f.ex. GMail
+- Malformed messages with incomplete headers will be propagated now
+- A notice is now emitted if the server does not support race-free Trash
+- Improved checking for invalid command lines
+- Options not supported due to the build configuration are still recognized
+  now, to make error messages more helpful
+- The progress indicator is rate-limited now
+- Various improvements to the debugging output
+- Vastly extended the autotest suite
+
+Bug Fixes:
 
-Changed default config & state locations to follow the XDG basedir spec.
-The old locations remain supported.
+- Worked around "unexpected EOF" error messages at end of TLS connections;
+  affects f.ex. GMail
+- Worked around protocol corruption issue with iCloud (mail.me.com)
+- Fixed missing CAPABILITY command after logging in if the server does not
+  report updated capabilities automatically (affects f.ex. MS Exchange)
+- Fixed CopyArrivalDate failing on some date strings
+- Fixed propagation of new messages to non-UIDPLUS servers
+- Fixed Timeout being ignored by DNS host resolution
+- Fixed broken Tunnel potentially causing SIGPIPE
+- Fixed Tunnel leaving behind zombie processes
+- Fixed expunges not being propagated at all if the first run after they
+  occurred did not include --delete
+- Fixed MaxMessages being exceeded when only --new was used
+- Fixed messages being instantly expired despite being important when only
+  --new was used
+- Trash-ing failures now prevent expunging and cause a non-zero exit code
+- Fixed placeholders being needlessly trashed
+- Fixed TrashNewOnly and TrashRemoteNew omitting messages for which only
+  a placeholder was synced
+- Fixed TrashRemoteNew omitting messages which exceed MaxSize
+- Fixed TrashRemoteNew not using race-free expunge
+- Optimized some places with unnecessarily high CPU usage
+- Fixed unnecessary network usage by non-selective uni-directional syncs when
+  no placeholders are present
+- Fixed crash when Patterns yields nothing when built with new compilers
+- Fixed crash when all flag propagations to a mailbox failed
+- Fixed handling of errors during opening mailboxes
+- Removed useless "lost track of ... messages" warnings when resuming after
+  an interruption
+- Fixed many minor bugs in corner cases, mostly when resuming after
+  interruptions
+
+
+1.4.4 (2021-12-03)
+==================
+
+Bug Fixes:
 
-The reference point for relative local paths in the configuration file
-is now the file's containing directory.
+- Fixed CVE-2021-3657: multiple buffer overflows on excessively large
+  IMAP literals
+- Fixed CVE-2021-44143: buffer overflow on header-less messages
+- Fixed crash on certain malformed messages in Maildirs
+- Fixed somewhat spurious notice about conflicting changes on
+  messages marked as deleted
 
-Placeholders will be now created for messages exceeding MaxSize even if
-they are flagged on the source side.
 
-Renamed the ReNew/--renew/-N options to Upgrade/--upgrade/-u
-and Delete/--delete/-d to Gone/--gone/-g.
+1.4.3 (2021-07-29)
+==================
 
-Superseded SSLVersions option with TLSVersions, and disabled TLS v1.0
-and v1.1 by default. Renamed SSLType option to TLSType.
+Improvements:
 
-Made the Channel side to expire with MaxMessages configurable.
+- Arbitrarily large GSSAPI authentication tokens are supported now
+- Symlink loops in Maildirs are detected now
 
-MaxMessages and MaxSize can be used together now.
 
-Added support for IMAP mailbox names with non-ASCII characters.
+1.4.2 (2021-06-06)
+==================
 
-Added support for Maildir Paths with suffixes.
+Improvements:
 
-The unfiltered list of mailboxes in each Store can be printed now.
+- INBOX miss-casing in Path setting is tolerated now
 
-A proper summary is now printed prior to exiting.
-This includes expunges, which are now included in the progress as well.
+Bug Fixes:
 
-Added new sync operation 'Old'.
+- Fixed CVE-2021-3578: possible remote code execution
+- Fixed crash on invalid CAPABILITY response code
 
-Added support for mirroring deletions more accurately.
 
-Added --dry-run option.
+1.3.6 (2021-06-06)
+==================
 
-Added --ext-exit option.
+Bug Fixes:
 
-[1.4.0]
+- Fixed CVE-2021-3578
 
-The 'isync' compatibility wrapper was removed.
 
-Added support for disabling TLS v1.3 - adjust SSLVersions if you set it.
-Removed support for obsolete/insecure SSL v3.
+1.4.1 (2021-02-21)
+==================
 
-The IMAP '$Forwarded' / Maildir 'P' (passed) flag is supported now.
+Bug Fixes:
 
-Support for configuring a TLS cipher string was added.
+- Fixed CVE-2021-20247: IMAP LIST/LSUB path traversal vulnerability
+- Fixed UIDVALIDITY change recovery potentially leading to data loss
+- Fixed spurious "unexpected FETCH response" errors with some servers
+- Worked around FastMail sending malformed PERMANENTFLAGS
 
-IMAP mailbox subscriptions are supported now.
 
-The IMAP user query can be scripted now.
+1.3.5 (2021-02-21)
+==================
 
-Added built-in support for macOS Keychain.
+Bug Fixes:
 
-Messages excluded by MaxSize will now result in placeholders.
+- Fixed CVE-2021-20247
 
-The use of Master/Slave terminology has been deprecated.
 
-[1.3.0]
+1.4.0 (2021-02-03)
+==================
 
-Network timeout handling has been added.
+Compatibility Concerns:
 
-Support for proper Maildir++ and a Maildir sub-folder naming style
-without extra dots have been added.
+- The 'isync' compatibility wrapper was removed
+- A C11 compiler is required for building now
+- The validity of the config file is checked more strictly now, including:
+  - Appearance of options in unexpected places
+  - The capitalization of INBOX
+- The new TLSv1.3 flag must be added to SSLVersions if the option is
+  used, unless disabling that version is desired (which is unlikely)
+- Removed support for the obsolete/insecure SSL v3
+- The use of Master/Slave terminology has been deprecated
 
-Support for TLS client certificates was added.
+New Features:
 
-Support for recovering from baseless UID validity changes was added.
+- The IMAP '$Forwarded' / Maildir 'P' (passed) flag is supported now
+- Support for configuring a TLS cipher string was added
+- IMAP mailbox subscriptions are supported now
+- The IMAP user query can be scripted now
+- Added built-in support for the macOS Keychain
+- Messages excluded by MaxSize will now result in placeholders
 
-The get-cert script was renamed to mbsync-get-cert.
+Bug Fixes:
 
-[1.2.0]
+- IMAP protocol errors are handled more robustly now
+- Fixed support for SASL's built-in EXTERNAL mechanism
+- Improved reliability of synchronization when resuming interrupted runs
+- Fixed MaxSize being ignored under certain circumstances when only one of
+  New and ReNew was requested
+- Fixed a network inefficiency occurring with server-side mailboxes that
+  receive new messages only via mbsync
 
-The 'isync' compatibility wrapper is now deprecated.
 
-An IMAP Path/NAMESPACE rooted in INBOX won't be handled specially any more.
-This means that some Patterns may need adjustment.
+1.3.4 (2021-02-03)
+==================
 
-The default output is a lot less verbose now.
-The meanings of the -V and -D options changed significantly.
+Bug Fixes:
 
-The SSL/TLS configuration has been re-designed.
-SSL is now explicitly enabled or disabled - "use SSL if available" is gone.
-Notice: Tunnels are assumed to be secure and thus default to no SSL.
+- Fixed regression in handling NAMESPACE 'INBOX.', introduced in v1.3.2
 
-Support for SASL (flexible authentication) has been added.
 
-Support for Windows file systems has been added.
+1.3.3 (2020-08-04)
+==================
 
-Support for compressed data transfer has been added.
+Improvements:
 
-Folder deletions can be propagated now.
+- PassCmd supports even bigger XOAUTH2 tokens now
 
-[1.1.0]
+Bug Fixes:
 
-Support for hierarchical mailboxes in Patterns.
+- Fixed crash on syncing multiple Channels which refer to different
+  Stores which use a common IMAPAccount
+- Fixed crash on IMAP connection breaking down while using -Dd
 
-Full support for IMAP pipelining (streaming, parallelization) added.
-This is considerably faster especially with high-latency networks.
 
-Faster and hopefully more reliable support for IMAP servers without the
-UIDPLUS extension (e.g., M$ Exchange).
+1.3.2 (2020-07-08)
+==================
 
-More automatic handling of SSL certificates.
+Improvements:
 
-IPv6 support.
+- Increased PassCmd buffer size to accommodate XOAUTH2 tokens
 
-IMAP password query can be scripted.
+Bug Fixes:
 
-Message arrival dates can be propagated.
+- Improved SSL error handling
+- Improved resilience to IMAP server bugs relating to FETCH
+- Fixed handling of non-uppercase IMAP iNbOx spellings
+- Fixed timeouts while uploading big messages
+- Fixed parsing of NIL hierarchy delimiters in IMAP LIST responses
+- Fixed crash when using Tunnel in an IPv6-enabled build
+- Fixed libcrypto detection from OpenSSL 1.1+ without pkg-config
+- Made --debug-crash work with Yama ptrace protection
 
-Data safety in case of system crashes was improved.
 
-MaxMessages was made vastly more useful.
+1.3.1 (2019-05-28)
+==================
 
-[1.0.0]
+Improvements:
 
-Essentially a rewrite. Synchronization state storage concept, configuration
-and command line changed entirely.
+- SSL now uses SNI, which for example GMail requires
+- The perl 5.14 requirement is now made explicit
+
+Bug Fixes:
+
+- Fixed fallbacks for missing UIDPLUS extension (with e.g. DavMail)
+- Fixed UIDVALIDITY recovery with really long Message-id headers
+- Fixed GSSAPI authentication with Kerberos
+- Fixed support for IMAP servers which do not sort search results (e.g.,
+  poczta.o2.pl)
+- Fixed CopyArrivalDate on platforms without glibc
+- Fixed useless SASL warnings with certain plugins
+- Improved OpenBSD support
+- Fixed a bunch of compiler warnings
+
+
+1.3.0 (2017-10-01)
+==================
+
+Compatibility Concerns:
+
+- If you have Maildir sub-folders, you need to update the configuration
+  to specify the naming style
+- get-cert was renamed to mbsync-get-cert to avoid namespace pollution
+- Attempts to enable SSL v2 now produce warnings. They were already
+  ineffective due to OpenSSL having removed the support a while ago.
+
+New Features:
+
+- Network timeout handling has been added
+- Support for proper Maildir++ and a Maildir sub-folder naming style
+  without extra dots have been added
+- Support for TLS client certificates was added
+- Support for recovering from baseless UID validity changes was added
+- The option DisableExtension for working around IMAP (server) bugs
+  was added
+
+Improvements:
+
+- Opening an IMAP Store is now immediately canceled when the opposite
+  Store failed to open
+
+Bug Fixes:
+
+- Fixed handling of UIDs upwards of 2^31
+- Fixed mailboxes being skipped with certain combinations of Patterns
+  and Path when the same IMAP Store is used in multiple Channels
+- Pattern INBOX* is not complained about any more if no Path is defined
+- Trashing messages now also resumes after an interruption, rather than
+  starting from scratch
+- Fixed spurious "TUID lost" warnings after interruption
+- Fixed various corner cases when resuming syncing after interruption.
+  The test suite exercises this much more thoroughly now.
+
+
+1.2.3 (2017-10-01)
+==================
+
+Improvements:
+
+- Enabled TLS 1.1 and 1.2 by default
+
+Bug Fixes:
+
+- Fixed cross-build with OpenSSL when using pkg-config
+- Fixed GCC 7 -Wimplicit-fallthrough warnings
+
+
+1.2.2 (2017-08-05)
+==================
+
+Improvements:
+
+- Added support for OpenSSL 1.1
+- IMAP UIDs up to 2^31 are now supported (as opposed to 10^9 before)
+
+Bug Fixes:
+
+- Fixed spurious reports about decompression errors
+- Fixed error reporting in both SSL server certificate validation
+  and IMAP COMPRESS support
+- Fixed use of IMAP LOGIN authentication in some configurations
+- Fixed handling of IMAP NAMESPACE without hierarchy delimiter
+- Fixed fallback to IPv4 when the kernel is built without IPv6 support
+
+
+1.2.1 (2015-11-08)
+==================
+
+Improvements:
+
+- The dependencies on Berkeley DB and zlib (and the features they enable)
+  can be explicitly disabled now
+- It is now possible to nest a Maildir Store's Path into its Inbox
+
+Bug Fixes:
+
+- Fixed IPv6-enabled builds crashing upon host name resolution failure
+- Fixed updating Maildir flags crashing on OpenBSD
+- Fixed deadlocks with NFS home directories
+- Fixed SASL authentication with GSS-API
+- Fixed messages being uploaded again and again with some servers
+- Fixed duplicate mailboxes with IMAP NAMESPACE "INBOX."
+- Fixed some problems in the IMAP command submission code
+- Fixed build with static libdb, libnsl, and libsocket
+- Fixed -DN not implying -Dn
+
+
+1.2.0 (2015-04-03)
+==================
+
+Compatibility concerns:
+
+- The 'isync' compatibility wrapper is now deprecated
+- Excess arguments in the config file are not silently ignored any more
+- An IMAP Path/NAMESPACE rooted in INBOX won't be handled specially
+  any more (the INBOX. prefix was never stripped).
+  This means that some Master/Slave/Patterns may need adjustment.
+- The SSL/TLS configuration has been re-designed.
+  SSL is now explicitly enabled or disabled - "use SSL if available" is gone.
+  Notice: Tunnels are assumed to be secure and thus default to no SSL.
+  The use of the system certificate store can be disabled now.
+- The default output is a lot less verbose now.
+  The meanings of the -V and -D options changed significantly.
+
+New Features:
+
+- Support for SASL (flexible authentication) has been added
+- Support for Windows file systems has been added by making a replacement
+  for the colon configurable
+- Support for compressed data transfer (IMAP COMPRESS) has been added
+- Folder deletions can be propagated now
+- The peak memory usage can be limited now
+
+Improvements:
+
+- The LITERAL+ IMAP extension will not be used any more when uploading big
+  messages, to enable early rejection of big uploads
+- A dysfunctional Store (e.g., bad password) will now be tried only once,
+  not for every Channel it is used in
+
+Bugfixes:
+
+- Fixed handling of some exceptional IMAP protocol conditions
+- Fixed spurious "unhandled SSL error 6" messages
+- Fixed failure to create Maildir Inbox
+
+
+1.1.3 (2015-04-03)
+==================
+
+Bug Fixes:
+
+- Fixed bogus "unexpected command continuation request"
+- Fixed out-of-Path INBOX never being matched by Patterns
+- Fixed escaping in PassCmd in mbsyncrc.sample
+
+
+1.1.2 (2015-01-18)
+==================
+
+Improvements:
+
+- Configurations with only TLS 1.1/1.2 are now actually possible
+- For security reasons, RequireSSL is not ignored anymore even when Tunnel
+  is used - you'll probably need to explicitly disable it
+- Maildir Stores with only Inbox but no Path are now permitted
+- IMAP Stores with an explicitly empty Path are now permitted
+
+Bug Fixes:
+
+- Fixed acceptance of trusted SSL host certificates
+- Fixed duplication of uploaded messages with some IMAP servers (in
+  particular, seznam.cz)
+- Fixed some memory management mistakes, some of which would cause crashes
+- Fixed garbage at end of folder names when using Patterns with some servers
+- Fixed Patterns misbehaving with certain server-provided namespaces
+
+
+1.1.1 (2014-06-01)
+==================
+
+Bug Fixes:
+
+- Fixed build on FreeBSD
+- Several fixes relating to badly handled IMAP responses,
+  including some crashes
+- Fixed folder handling in isync compatibility wrapper
+- Added pedantic error handling in some rather unlikely places
+- Man page fixes
+- Some cosmetical fixes
+
+
+1.1.0 (2013-12-18)
+==================
+
+New Features:
+
+- Added support for hierarchical mailboxes in Patterns
+- Added full support for IMAP pipelining (streaming, parallelization).
+  This is considerably faster especially with high-latency networks.
+- Added IPv6 support
+- IMAP password query can be scripted now; option PassCmd
+- Message arrival dates can be propagated now; option CopyArrivalDate
+- Added ExpireUnread option
+
+Improvements:
+
+- Data safety in case of system crashes was improved
+- Faster and hopefully more reliable support for IMAP servers without
+  the UIDPLUS extension (e.g., M$ Exchange)
+- More automatic handling of SSL certificates
+- MaxMessages was made vastly more useful
+
+Bug Fixes:
+
+(rumored)
+
+
+1.0.6 (2013-02-20)
+==================
+
+Bug Fixes:
+
+- Fixed CVE-2013-0289: missing SSL subject verification
+- Fixed 64-bit cleanness
+
+
+1.0.5 (2012-04-28)
+==================
+
+Improvements:
+
+- In the wrapper, give the implicitly created IMAP Account config the
+  name of the Store
+
+Bug Fixes:
+
+- Fixed crash if neither Host nor Tunnel are specified
+- Fixed handling of failure to store messages
+- Fixed hang after failed start_tls
+- Fixed memory access error (used memcpy for overlapping regions)
+
+
+1.0.4 (2008-02-23)
+==================
+
+- Bug Fixes
+
+
+1.0.3 (2006-11-03)
+==================
+
+- Bug Fixes
+
+
+1.0.2 (2006-02-25)
+==================
+
+- Bug Fixes
+
+
+1.0.1 (2005-03-28)
+==================
+
+- Bug Fixes
+
+
+1.0.0 (2004-09-15)
+==================
+
+This is essentially a rewrite. Synchronization state storage concept,
+configuration and command line changed entirely.
 But you needn't to worry about the upgrade, as a fully automated migration
 path is provided, even for users of isync 0.7 and below.
-Still, you should re-read the manual to be able to take full advantage of the
-new features:
+Still, you should re-read the manual to be able to take full advantage of
+the new features:
+
+- The supported mailbox types can be freely paired now.
+  A possible application of this is using a local IMAP server to access
+  mailboxes that are not natively supported yet.
+- Message deletions (expunges) are now propagated both ways, so there is
+  no need for using mutt with maildir_trash any more
+- Additional trash options added
+- 'OneToOne' was replaced by something more flexible
+- Added partial support for IMAP pipelining (streaming, parallelization).
+  This makes flag change propagation much faster - this affects every
+  message that becomes Seen/Read.
+
+
+0.9.2 (2003-12-07)
+==================
+
+- Bug Fixes
+
+
+0.9.1 (2003-05-07)
+==================
+
+- Bug Fixes
+
+
+0.9 (2003-05-05)
+================
 
-The supported mailbox types can be freely paired.
-A possible application of this is using a local IMAP server to access
-mailboxes that are not natively supported yet.
+Compatibility Concerns:
 
-Message deletions (expunges) are now propagated both ways, so there is no need
-for using mutt with maildir_trash any more.
+- -C now creates both local and remote boxes; the new -L and -R create
+  only local/remote ones.
 
-Additional trash options added.
+New Features:
 
-`OneToOne' replaced by something more flexible.
+- Added Tunnel directive to allow the user to specify a shell command to
+  run to set up an IMAP connection in place of a TCP socket (e.g., to run
+  over an SSH session)
+- Added PREAUTH support (useful mostly in conjunction with Tunnel)
+- Added 'OneToOne' configuration option: ignore any Mailbox specifications
+  and instead pick up all mailboxes from the local MailDir and remote
+  Folder, and map them 1:1 onto each other according to their names
 
-Partial support for IMAP pipelining (streaming, parallelization) added.
-Makes flag change propagation much faster - this affects every message that
-becomes Seen/Read.
+Improvements:
 
-[0.9]
+- Messages marked deleted are not uploaded when we are going to expunge
+- Locally generated messages are not re-fetched after uploading even if
+  the UIDPLUS extension is not supported by the server
+- --quiet is now really quiet
 
-Added Tunnel directive to allow the user to specify a shell command to run
-to set up an IMAP connection in place of a TCP socket (eg., to run over
-an SSH session).
 
-Added PREAUTH support (useful mostly in conjunction with Tunnel).
+0.8 (2002-01-28)
+================
 
-Messages marked deleted are not uploaded when we are going to expunge.
+Compatibility Concerns:
 
-Locally generated messages are not re-fetched after uploading even if the
-UIDPLUS extension is not supported by the server.
+- In order to fix the problem where messages copied from one mailbox to
+  another were not uploaded to the new mailbox, the way Isync stores the
+  UID for each message needed to be changed.  As a result, you _MUST_
+  delete all the messages in the local maildir box before using this
+  version. Otherwise it will upload every message to the server thinking
+  it is a new mail.
 
-Added `OneToOne' configuration option: ignore any Mailbox specifications
-and instead pick up all mailboxes from the local MailDir and remote Folder
-and map them 1:1 onto each other according to their names.
 
--C now creates both local and remote boxes; -L and -R create only local/remote.
+0.7 (2001-11-20)
+================
 
---quiet is now really quiet.
+New Features:
 
-[0.8]
+- Added 'MaxMessages' configuration option to allow tracking of only the
+  most recently added message in the local mailbox
+- Added --create (-C) command line option to force creation of the local
+  maildir-style mailbox if it doesn't already exist
 
-!!! IMPORTANT !!!
 
-In order to fix the problem where messages copied from one mailbox to
-another were not uploaded to the new mailbox, the way Isync stores the UID
-for each message needed to be changed.  As a result, you _MUST_ delete all
-the messages in the local maildir box before using this version.  Otherwise
-it will upload every message to the server thinking its a new mail.
+0.6 (2001-10-31)
+================
 
-[0.7]
+New Features:
 
-Added `MaxMessages' configuration option to allow tracking of only the most
-recently added message in the local mailbox.
+- Added 'Delete' configuration option to correspond to the -d command
+  line option
+- Added -a (--all) command line option to synchronize all mailboxes
 
-Added --create (-C) command line option to force creation of the local
-maildir-style mailbox if it doesn't already exist.
 
-[0.6]
+0.5 (2001-06-13)
+================
 
-Added `Delete' configuration option to correspond to the -d command line
-option.
+New Features:
 
-Added -a (--all) command line option to synchronize all mailboxes.
+- Updated SSL support
+- Added CRAM authentication support
+- Added MailDir configuration option to specify the default location of
+  local mailboxes when relative paths are used
+- Added support for uploading local messages to the IMAP server
+- Added CopyDeletedTo configuration option to cause isync to move deleted
+  messages to a particular mailbox on the server when they are expunged
 
-[0.5]
 
-Updated SSL support.
+0.4 (2000-12-31)
+================
 
-Added CRAM authentication support.
+New Features:
 
-Added MailDir configuration option to specify the default location of local
-mailboxes when relative paths are used.
+- Added MaxSize configuration option to limit downloading of new messages
+  from the server to less than some threshold
 
-Added support for uploading local messages to the IMAP server.
+Improvements:
 
-Added CopyDeletedTo configuration option to cause isync to move deleted
-messages to a particular mailbox on the server when they are expunged.
+- More robust --fast option works without using \Recent flags, so the
+  previous problem with multiple accesses killing these flags is no longer
+  a problem
+- RFC2060 obsoleted RFC822.PEEK; use BODY.PEEK[] instead which does the
+  same job
+- Stop requesting UID in FETCH when doing UID FETCH (RFC2060 states that
+  it is automatically returned)
 
-[0.4]
 
-Added MaxSize configuration option to limit downloading of new messages from
-the server to less than some threshold.
+0.3 (2000-12-21)
+================
 
-More robust --fast option works without using \Recent flags, so the previous
-problem with multiple accesses killing these flags is no longer a problem.
+New Features:
 
-RFC2060 obsoleted RFC822.PEEK, use BODY.PEEK[] instead which does the same
-job.
+- Added support for building RPMs
 
-Don't need to request UID in a FETCH when doing UID FETCH (RFC2060 states
-that its automatically returned).
+Bug Fixes:
 
-[0.3]
+- Fixed failure to clean up temp maildir files when the fetch of a new
+  message failed
+- Fixed invalid assumption of order of the flags returned by "UID FETCH"
 
-Fixed to clean up temp maildir files when the fetch of a new message failed.
 
-Fixed to not assume order of the flags returned by "UID FETCH"
+0.2 (2000-12-21)
+================
 
-Added support for building RPMs.
+New Features:
 
-[0.2]
+- Added SSL support
 
-SSL support added.
 
-[0.1]
+0.1 (2000-12-20)
+================
 
-Initial release.
+- Initial release

[commit] master: stub out ChangeLog

[ossi <os...@us...>]

commit 8f58baa153fc8bbed2e44f993e13a364a2ab4ac4
Author: Oswald Buddenhagen <os...@us...>
Date:   Tue Mar 11 19:05:05 2025 +0100

    stub out ChangeLog
    
    that file is a relic from the times before distributed VCS with atomic
    commits existed. ours was slightly filtered, but still way too noisy to
    be useful for end users. so basically, it had no target audience.
    
    the file itself is left in place, because automake complains otherwise
    (we don't want to use 'foreign' strictness, as otherwise INSTALL is not
    added by autoreconf --install).

 .gitignore  |  1 -
 ChangeLog   |  3 +++
 Makefile.am | 55 -----------------------------------------------------
 autogen.sh  |  1 -
 4 files changed, 3 insertions(+), 57 deletions(-)

diff --git a/.gitignore b/.gitignore
index 66a12ec..31ee130 100644
--- a/.gitignore
+++ b/.gitignore
@@ -1,5 +1,4 @@
 /.autoconf_trace
-/ChangeLog
 /INSTALL
 /VERSION
 /autom4te.cache/
diff --git a/ChangeLog b/ChangeLog
new file mode 100644
index 0000000..5fc59d9
--- /dev/null
+++ b/ChangeLog
@@ -0,0 +1,3 @@
+This file is empty for a lack of a target audience.
+
+Have a look at the NEWS file for a summary of user-visible changes.
diff --git a/Makefile.am b/Makefile.am
index 79067cb..121ca3e 100644
--- a/Makefile.am
+++ b/Makefile.am
@@ -6,61 +6,6 @@ SUBDIRS = src
 bin_SCRIPTS = mbsync-get-cert
 EXTRA_DIST = LICENSES VERSION debian isync.spec $(bin_SCRIPTS)
 
-LOG_PL = \
-    use POSIX qw(strftime); \
-    use Date::Parse; \
-    use Text::Wrap; \
-    $$Text::Wrap::columns = 72; \
-    while (defined($$_ = <>)) { \
-        /^commit / or die "commit missing: $$_"; \
-        <> =~ /^log size (\d+)$$/ or die "wrong size"; \
-        $$len = $$1; \
-        read(STDIN, $$log, $$len) == $$len or die "unexpected EOF"; \
-        $$log =~ s/^Author: ([^>]+>)\nDate:   (\d{4}-\d\d-\d\d \d\d:\d\d:\d\d [-+]\d{4})\n(.*)$$/$$3/s or die "unexpected log format"; \
-        $$author = $$1; $$date = str2time($$2); \
-        scalar(<>); \
-        @files = (); \
-        $$pfx = ""; \
-        while (defined($$l = <>) and $$l ne "\n") { \
-            chomp $$l; \
-            next if ($$l =~ m,^(ChangeLog$$|NEWS$$|TODO$$|debian/),); \
-            if (!@files) { \
-                $$pfx = $$l; \
-                $$pfx =~ s,/?[^/]+$$,,; \
-            } else { \
-                while (length($$pfx)) { \
-                    $$l =~ m,^\Q$$pfx/\E, and last; \
-                    $$pfx =~ s,/?[^/]+$$,,; \
-                } \
-            } \
-            push @files, $$l; \
-        } \
-        next if (!@files); \
-        print strftime("%F %H:%M", gmtime($$date))."  ".$$author."\n\n"; \
-        if (@files > 1 and ($$len = length($$pfx))) { \
-            @efiles = (); \
-            for $$f (@files) { push @efiles, substr($$f, $$len + 1); } \
-            $$fstr = $$pfx."/: "; \
-        } else { \
-            @efiles = @files; \
-            $$fstr = ""; \
-        } \
-        print wrap("\t* ", "\t  ", $$fstr.join(", ", @efiles).":")."\n"; \
-        $$log =~ s, +$$,,gm; \
-        $$log =~ s,^    ,\t,gm; \
-        print $$log."\n"; \
-    }
-
-$(srcdir)/.git/index:
-$(srcdir)/ChangeLog: $(srcdir)/.git/index
-	$(MAKE) log
-
-log:
-	@test -z "$(srcdir)" || cd $(srcdir) && \
-	 ( ! test -d .git || \
-	   git log --pretty=medium --date=iso --log-size --name-only --no-merges | \
-	     perl -e '$(LOG_PL)' > ChangeLog )
-
 cov-scan: clean
 	/opt/cov-analysis-*/bin/cov-build --dir cov-int $(MAKE)
 	tar cavf isync-cov.tar.xz cov-int
diff --git a/autogen.sh b/autogen.sh
index bd877a4..86518ed 100755
--- a/autogen.sh
+++ b/autogen.sh
@@ -1,4 +1,3 @@
 #! /bin/sh
 set -e -v
-make -f Makefile.am log
 autoreconf -f -i

[commit] master: prefix deprecation notices with file name

[ossi <os...@us...>]

commit 281a9ba465ba8be0d58fcb690ec34f9cdacdd1a9
Author: Oswald Buddenhagen <os...@us...>
Date:   Tue Mar 11 12:44:22 2025 +0100

    prefix deprecation notices with file name
    
    so it's (even) more obvious that these refer to the configuration,
    rather than being some internal sloppiness.

 src/config.c   | 6 +++---
 src/drv_imap.c | 4 ++--
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/src/config.c b/src/config.c
index 4ff389e..3ea4837 100644
--- a/src/config.c
+++ b/src/config.c
@@ -660,11 +660,11 @@ load_config( const char *where )
 	}
 	fclose (cfile.fp);
 	if (cfile.ms_warn)
-		warn( "Notice: Master/Slave are deprecated; use Far/Near instead.\n" );
+		warn( "%s: notice: Master/Slave are deprecated; use Far/Near instead.\n", cfile.file );
 	if (cfile.renew_warn)
-		warn( "Notice: ReNew is deprecated; use Upgrade instead.\n" );
+		warn( "%s: notice: ReNew is deprecated; use Upgrade instead.\n", cfile.file );
 	if (cfile.delete_warn)
-		warn( "Notice: Delete is deprecated; use Gone instead.\n" );
+		warn( "%s: notice: Delete is deprecated; use Gone instead.\n", cfile.file );
 	cfile.err |= merge_ops( gcops, global_conf.ops, "" );
 	if (!global_conf.sync_state) {
 		const char *state_home = getenv( "XDG_STATE_HOME" );
diff --git a/src/drv_imap.c b/src/drv_imap.c
index b0be1fd..9b170ab 100644
--- a/src/drv_imap.c
+++ b/src/drv_imap.c
@@ -3824,7 +3824,7 @@ imap_parse_store( conffile_t *cfg, store_conf_t **storep )
 			static int sslv_warned;
 			if (!sslv_warned) {
 				sslv_warned = 1;
-				warn( "Notice: SSLVersions is deprecated. Use TLSVersions instead.\n" );
+				warn( "%s: notice: SSLVersions is deprecated. Use TLSVersions instead.\n", cfg->file );
 			}
 			server->sconf.ssl_versions = 0;
 			arg = cfg->val;
@@ -3864,7 +3864,7 @@ imap_parse_store( conffile_t *cfg, store_conf_t **storep )
 			static int sslt_warned;
 			if (!sslt_warned) {
 				sslt_warned = 1;
-				warn( "Notice: SSLType is deprecated. Use TLSType instead.\n" );
+				warn( "%s: notice: SSLType is deprecated. Use TLSType instead.\n", cfg->file );
 			}
 		  tlstype:
 			if (!strcasecmp( "None", cfg->val )) {

Re: Contention between mbsync clients

[Liam H. <li...@hp...>]

Oswald Buddenhagen <osw...@gm...> writes:

> On Mon, Nov 18, 2024 at 09:51:53PM -0600, Liam Hupfer wrote:
>>now my desktop seems to move them back when syncing.
>>
> no idea why it would do that.
> add -Ds to the command line and try to make sense of what it says.

Figured it out. Nothing to do with mbsync, really. Thanks for your help!

—

For posterity’s sake, it was a shoddy notmuch/afew configuration. I was
attempting to synchronize maildir subfolders with notmuch tags via
afew’s [move mode]. The notmuch database uses the Message-ID as a key and
only marks new mail if the Message-ID is new. If mail moved on the far
side (i.e. I triaged mail with another client), mbsync would sync the
moves correctly, but notmuch would not consider the moved mail as new,
so `afew --tag --new' would not update the folder-based tags. Then `afew
--move-mails' would move the mail /back/ to the existing, outdated
folder tag from the notmuch database. And `mbsync' would of course
diligently sync these changes, appearing to me as “contention”.

The moral of the story is I need to bid adieu to IMAP folders and move
to JMAP.

—Liam


[move mode] <https://afew.readthedocs.io/en/latest/commandline.html#move-mode>

[commit] master: fix -ls hanging after synchronous error

[ossi <os...@us...>]

commit 61f0810bb17ba185aa12fd2197f92deaee249dbd
Author: Oswald Buddenhagen <os...@us...>
Date:   Tue Jan 28 10:30:47 2025 +0100

    fix -ls hanging after synchronous error
    
    we need to make a note that we're already done with the Store, as
    otherwise we'll wait for a callback that never comes.

 src/main_list.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/src/main_list.c b/src/main_list.c
index c4aa63e..ff3d193 100644
--- a/src/main_list.c
+++ b/src/main_list.c
@@ -78,6 +78,7 @@ list_store_bad( void *aux )
 {
 	list_vars_t *lvars = (list_vars_t *)aux;
 
+	lvars->done = 1;
 	lvars->drv->cancel_store( lvars->ctx );
 	lvars->cvars->ret = 1;
 	list_next_store( lvars );

[commit] master: fix --dry-run without --debug-driver

[ossi <os...@us...>]

commit 1e0b661d09c5a4bbb0bb2061db8071363145d5ac
Author: Oswald Buddenhagen <os...@us...>
Date:   Tue Jan 28 00:49:17 2025 +0100

    fix --dry-run without --debug-driver
    
    the stubbing is implemented by the proxy driver, so we need to hook it
    in in that case as well. the driver is already prepared for that, as
    it's also used to implement -Ta/-TA.

 src/main_sync.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/src/main_sync.c b/src/main_sync.c
index 2ea4a92..f4ec34b 100644
--- a/src/main_sync.c
+++ b/src/main_sync.c
@@ -450,7 +450,7 @@ do_sync_chans( main_vars_t *mvars )
 			labels[F] = labels[N] = "";
 		for (int t = 0; t < 2; t++) {
 			store_t *ctx = mvars->drv[t]->alloc_store( mvars->chan->stores[t], labels[t] );
-			if ((DFlags & DEBUG_DRV) || ((DFlags & FORCEASYNC(t)) && !(dcaps[t] & DRV_ASYNC))) {
+			if ((DFlags & (DEBUG_DRV | DRYRUN)) || ((DFlags & FORCEASYNC(t)) && !(dcaps[t] & DRV_ASYNC))) {
 				mvars->drv[t] = &proxy_driver;
 				ctx = proxy_alloc_store( ctx, labels[t], DFlags & FORCEASYNC(t) );
 			}

Re: Does isync report an error in case of data corruption in RAM?

[Oswald B. <osw...@gm...>]

On Mon, Jan 13, 2025 at 04:33:32PM -0800, Hong Xu wrote:
>However, consumer-grade hardware is prone to RAM bit flips, at a higher
>rate than what most people probably think. The most recent research I
>knew is discussed on Wikipedia [2], which shows the error rate was on
>the magnitude of 1 bit error per gigabyte of RAM per 1.8 hours.
>
did you actually read that google paper?
as far as i'm concerned, its conclusion is "bad ram is bad, duh".
a subsequent study by facebook
(https://www.dpss.inesc-id.pt/~romanop/files/secMemErr.pdf) drives the
point home even harder.
it's not a non-issue, but it's by far not as dramatic as the number
above suggests, and running a ram test on a regular basis should reveal
most problems before they cause much damage.

Re: Does isync report an error in case of data corruption in RAM?

[Hong Xu <ho...@to...>]

On 2025-01-13 Mon 04:25 GMT-08, Oswald Buddenhagen <osw...@gm...> wrote:

> On Mon, Jan 13, 2025 at 01:33:54AM -0800, Hong Xu wrote:
>>Does isync perform any checks (like some checksum) to prevent data loss
>>in an event of RAM error?
>>
> no, and the idea doesn't even make much sense. there would have to be a
> checksum to compare to, and that couldn't be calculated before the data
> was actually loaded into ram.

I did a bit of research into IMAP, and I agree that there's no way to
fully fix this unless there are data integrity support at the IMAP
level.

>
> the whole idea of software-based ram-checking would make sense only for
> data that lives in ram for a long time, where corruption due to storage
> deterioration would be much more likely than corruption during transfer,
> or the added time due to computing the checksums at storage and
> retrieval.
>
> isync stores message payloads only for a few seconds in its network
> buffers. if you are concerned about integrity at that time scale, get
> hardware that can do it for you automatically. it would be total
> insanity to half-assedly harden each application.

Even though I don't think isync can fully fix the issue here, I'd still
like to clarify the concern for isync users, which is real and shouldn't
be dismissed. I'd also like to share what I do to mitigate.

=== Concern ===

With proper backup, storage corruption is not a concern here, because
storage corruptions can hardly go undetected due to the builtin error
checks in hard drives, even on consumer-grade hardware. When a
corruption occurs, data can be read without error only at a super-low
probability. See [1] and its reference.

However, consumer-grade hardware is prone to RAM bit flips, at a higher
rate than what most people probably think. The most recent research I
knew is discussed on Wikipedia [2], which shows the error rate was on
the magnitude of 1 bit error per gigabyte of RAM per 1.8 hours.

And they truly affect data integrity, which is a frequent topic
discussed in the NAS community for one example, because data can go
corrupted in RAM in the period of being copied from one location to
another (e.g., [3]).

=== My Solution ===

I'd also like to share what I do to mitigate silent data corruption when
it's impractical to use an ECC RAM:

- Regularly back up email folders. This can become a life-saver if a RAM
  becomes faulty, in which case many file corruptions and system crashes
  becomes noticeable. I can then restore the backups and let them sync
  back to the server with isync.

- Save all important data, like attachment, to somewhere else with data
  integrity checks, such as a local git repository, or a subversion
  repository if there are large binaries. In this way, I don't worry
  about silent data corruptions when copying them around in the future,
  and occasional bit flip in emails becomes non-concerning.

[1] https://serverfault.com/a/77721/158338
[2] https://en.wikipedia.org/wiki/ECC_memory#Research
[3] https://forums.truenas.com/t/ecc-memory-for-home-servers-or-not/7099



-- 
Hong

Re: Does isync report an error in case of data corruption in RAM?

[Oswald B. <osw...@gm...>]

On Mon, Jan 13, 2025 at 01:33:54AM -0800, Hong Xu wrote:
>Does isync perform any checks (like some checksum) to prevent data loss
>in an event of RAM error?
>
no, and the idea doesn't even make much sense. there would have to be a
checksum to compare to, and that couldn't be calculated before the data
was actually loaded into ram.
the whole idea of software-based ram-checking would make sense only for
data that lives in ram for a long time, where corruption due to storage
deterioration would be much more likely than corruption during transfer,
or the added time due to computing the checksums at storage and
retrieval.
isync stores message payloads only for a few seconds in its network
buffers. if you are concerned about integrity at that time scale, get
hardware that can do it for you automatically. it would be total
insanity to half-assedly harden each application.

[commit] master: explicitly strip newline from contents of VERSION

[ossi <os...@us...>]

commit 277e3cee1b5c31e221b2bfd4592d09c41a9a2226
Author: Oswald Buddenhagen <os...@us...>
Date:   Mon Jan 13 13:09:18 2025 +0100

    explicitly strip newline from contents of VERSION
    
    while m4 1.4.19 + autoconf 2.72 from my debian unstable drop the
    trailing newline automatically, m4 1.4.18 + autoconf 2.69 shipped
    with ubuntu 20.4 don't, leading to an invalid configure script.
    
    reported and verified by Donat Wegner <dm...@we...>.

 configure.ac | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/configure.ac b/configure.ac
index 7f1fc6e..8c46b8b 100644
--- a/configure.ac
+++ b/configure.ac
@@ -3,7 +3,7 @@
 dnl SPDX-License-Identifier: GPL-2.0-or-later
 
 m4_syscmd([./version.sh])
-AC_INIT([isync], m4_sinclude([VERSION]))
+AC_INIT([isync], m4_chomp(m4_sinclude([VERSION])))
 AC_CONFIG_HEADERS([autodefs.h])
 
 AC_CANONICAL_TARGET

Does isync report an error in case of data corruption in RAM?

[Hong Xu <ho...@to...>]

Hi all,

I'm interested in understanding whether isync is capable of preventing
data loss in case of RAM errors.

Bit corruption in RAM is a low-probability event but does happen. ECC
memories [1] exist to address this concern. However, today, most
personal computers use non-ECC memories and are prone to such
corruption. [2]

In the context of email syncing, this means that some silent data
corruption may occur if there's no integrity check. For example, a
message is synced to local, but its content can become corrupted in RAM
before written to disk.

Another consequence may occur if the RAM starts to be faulty. An ECC RAM
will refuse to work at all if it's too faulty, preventing data loss. But
a non-ECC RAM can just corrupt whatever is being read from the RAM and
written to the disk at the moment.

Does isync perform any checks (like some checksum) to prevent data loss
in an event of RAM error?

Thank you!


[1] https://en.wikipedia.org/wiki/ECC_memory
[2] You may have heard that all DDR5 RAMs have "ECC", but this is untrue. See https://superuser.com/a/1797176/139328
-- 
Hong

Re: synchronize Opentext/Groupwise Notes+Appointments?

[Peter P. <pet...@fa...>]

Thanks for the reply Oswald!

Meanwhile I found out that these "Notes" and "Appointments" are only
displayed in the webmail Inbox, but are in fact stored in a separate
imap folder which I wasn't syncing so far. I am doing this now and
mbsync gracefully syncs these objects as well. Sorry for the noise and
thanks again!

Peter

* Oswald Buddenhagen via isync-devel <isy...@li...> [2024-12-20 11:03]:
> On Fri, Dec 20, 2024 at 10:08:29AM +0100, Peter P. wrote:
> > Can someone possibly confirm that these
> > messages are not transferred by mbsync from an imap server?
> > 
> isync transfers everything it sees on imap, and that doesn't even have
> to be messages (though objects not compliant with rfc822 would get
> garbled). you can use another imap client (say, thunderbird) to confirm
> that isync isn't just blind. maybe these specials are virtually mapped
> to separate mailboxes?
>
> obviously, nobody here knows anything about this. have you googled
> and/or asked in a forum related to that server?

Re: synchronize Opentext/Groupwise Notes+Appointments?

[Oswald B. <osw...@gm...>]

On Fri, Dec 20, 2024 at 10:08:29AM +0100, Peter P. wrote:
>Can someone possibly confirm that these
>messages are not transferred by mbsync from an imap server?
>
isync transfers everything it sees on imap, and that doesn't even have
to be messages (though objects not compliant with rfc822 would get
garbled). you can use another imap client (say, thunderbird) to confirm
that isync isn't just blind. maybe these specials are virtually mapped
to separate mailboxes?

obviously, nobody here knows anything about this. have you googled
and/or asked in a forum related to that server?

Re: synchronize Opentext/Groupwise Notes+Appointments?

[Peter P. <pet...@fa...>]

Dear list,

still trying to figure out how to sync the aforementioned "Notes" and
"Appointments" using mbsync. Can someone possibly confirm that these
messages are not transferred by mbsync from an imap server?

Thanks again,
Peter

* Peter P. <pet...@fa...> [2024-12-15 06:57]:
> Hi list,
> 
> I discovered that people are from time to time sending me "Notes"
> instead of emails, or "Appointments". These seem to be something
> different and are not synced by mbsync, making me miss messages and
> appointments.
> 
> The anatomy of such an "appointment" is:
> MESSAGE 	2 KB 
> TEXT.htm 	6 KB 	
> rfc2445.ics 	7 KB 	
> Mime.822 	40 KB 
> 
> and of a "Note":
> MESSAGE 	1 KB
> text.htm 	1 KB 
> 
> Is there a way to sync them, or at least convert them on the server to
> something that can be synced?
> 
> Thanks!
> Peter

Re: config working on Ubuntu not working an arch?

[Bence F. <be...@fe...>]

On Thu Dec 19, 2024 at 10:35, Tassilo Horn <ts...@gn...> wrote:
> "Bence Ferdinandy" <be...@fe...> writes:
>
> Hi Bence,
>
>> My previous machine was running Ubuntu now I moved to CachyOS (arch
>> btw).  I supposedly didn't change anything in my configs and I can
>> actually send messages with msmtp for my oauth accounts (gmail and
>> outlook), but mbsync if failing for both.
>>
>> For example with `mbsync -D`:
>>
>>     Connecting to imap.gmail.com (173.194.76.109:993)...
>>     Connection is now encrypted
>>     F: * OK Gimap ready for requests from 89.223.195.239 o37mb130304551wms
>>     F: >>> 1 CAPABILITY
>>     F: * CAPABILITY IMAP4rev1 UNSELECT IDLE NAMESPACE QUOTA ID XLIST CHILDREN
>> X-GM-EXT-1 XYZZY SASL-IR AUTH=XOAUTH2 AUTH=PLAIN AUTH=PLAIN-CLIENTTOKEN
>> AUTH=OAUTHBEARER
>>     F: 1 OK Thats all she wrote! o37mb130304551wms
>>     Logging in...
>>     Authenticating with SASL mechanism PLAIN...
>>     F: >>> 2 AUTHENTICATE PLAIN <authdata>
>>     F: 2 NO [AUTHENTICATIONFAILED] Invalid credentials (Failure)
>>     IMAP command 'AUTHENTICATE PLAIN <authdata>' returned an error:
>> [AUTHENTICATIONFAILED] Invalid credentials (Failure)
>
> Hm, I use Arch with the extra/isync package which is currently version
> 1.5.0-2 and it works for me (on the AUR there's only an isync-git
> package).  That's my debug log for my gmail account:
>
> --8<---------------cut here---------------start------------->8---
> Connecting to imap.gmail.com ([2a00:1450:400c:c00::6c]:993)... 
> Connection is now encrypted
> F: * OK Gimap ready for requests from 2003:df:1703:d00:5e51:4fff:fef3:ece7 b9mb127754308wmb
> F: >>> 1 CAPABILITY
> F: * CAPABILITY IMAP4rev1 UNSELECT IDLE NAMESPACE QUOTA ID XLIST CHILDREN X-GM-EXT-1 XYZZY SASL-IR AUTH=XOAUTH2 AUTH=PLAIN AUTH=PLAIN-CLIENTTOKEN AUTH=OAUTHBEARER
> F: 1 OK Thats all she wrote! b9mb127754308wmb
> Logging in...
> Authenticating with SASL mechanism PLAIN...
> F: >>> 2 AUTHENTICATE PLAIN <authdata>
> F: * CAPABILITY IMAP4rev1 UNSELECT IDLE NAMESPACE QUOTA ID XLIST CHILDREN X-GM-EXT-1 UIDPLUS COMPRESS=DEFLATE ENABLE MOVE CONDSTORE ESEARCH UTF8=ACCEPT LIST-EXTENDED LIST-STATUS LITERAL- SPECIAL-USE APPENDLIMIT=35651584
> F: 2 OK ts...@gm... authenticated (Success)
> F: >>> 3 COMPRESS DEFLATE
> F: 3 OK Success
> F: >>> 4 ENABLE UTF8=ACCEPT
> F: * ENABLED UTF8=ACCEPT
> F: 4 OK Success
> F: >>> 5 NAMESPACE
> F: * NAMESPACE (("" "/")) NIL NIL
> F: 5 OK Success
> --8<---------------cut here---------------end--------------->8---
>
> However, I don't use OAuth2 for authentication but an app password.  It
> seems the plugin you've built yourself on Ubuntu is available on the
> AUR:
>
> --8<---------------cut here---------------start------------->8---
> ❯ yay -s xoauth2                                                                        
> 3 aur/sasl-xoauth2-git r203.47ff232-2 (+0 0.00) 
>     SASL plugin that enables client-side use of OAuth 2.0
> 2 aur/oauth2ms-git r10.a1ef0cabfdea-1 (+0 0.00) (Orphaned) 
>     XOAUTH2 compatible O365 token fetcher
> 1 aur/cyrus-sasl-xoauth2-git r24.36aabca54fd6-1 (+5 0.13) 
>     XOAUTH2 mechanism plugin for cyrus-sasl
> --8<---------------cut here---------------end--------------->8---
>
> Maybe try installing that.

Thanks everybody for the pointers, one step closer now :)


Logging in...
Authenticating with SASL mechanism XOAUTH2...
Error performing SASL authentication step: SASL(-1): generic failure: Unable to find a callback: 32775

❯ paru -Q | grep sasl
cyrus-sasl 2.1.28-4.1
cyrus-sasl-xoauth2-git r24.36aabca54fd6-1
gsasl 2.2.1-1.1
libsasl 2.1.28-5.1
sasl-xoauth2-git r206.bb5c890-1

Tried building isync also, but that has the same output. I also restarted, just
to make sure.

Nevermind, after uninstalling `sasl-xoauth2-git` it works (thanks Marci!). So tldr:

these two packages are needed on arch:
cyrus-sasl 
cyrus-sasl-xoauth2-git 

Best,
Bence

Re: config working on Ubuntu not working an arch?

[Tassilo H. <ts...@gn...>]

"Bence Ferdinandy" <be...@fe...> writes:

Hi Bence,

> My previous machine was running Ubuntu now I moved to CachyOS (arch
> btw).  I supposedly didn't change anything in my configs and I can
> actually send messages with msmtp for my oauth accounts (gmail and
> outlook), but mbsync if failing for both.
>
> For example with `mbsync -D`:
>
>     Connecting to imap.gmail.com (173.194.76.109:993)...
>     Connection is now encrypted
>     F: * OK Gimap ready for requests from 89.223.195.239 o37mb130304551wms
>     F: >>> 1 CAPABILITY
>     F: * CAPABILITY IMAP4rev1 UNSELECT IDLE NAMESPACE QUOTA ID XLIST CHILDREN
> X-GM-EXT-1 XYZZY SASL-IR AUTH=XOAUTH2 AUTH=PLAIN AUTH=PLAIN-CLIENTTOKEN
> AUTH=OAUTHBEARER
>     F: 1 OK Thats all she wrote! o37mb130304551wms
>     Logging in...
>     Authenticating with SASL mechanism PLAIN...
>     F: >>> 2 AUTHENTICATE PLAIN <authdata>
>     F: 2 NO [AUTHENTICATIONFAILED] Invalid credentials (Failure)
>     IMAP command 'AUTHENTICATE PLAIN <authdata>' returned an error:
> [AUTHENTICATIONFAILED] Invalid credentials (Failure)

Hm, I use Arch with the extra/isync package which is currently version
1.5.0-2 and it works for me (on the AUR there's only an isync-git
package).  That's my debug log for my gmail account:

--8<---------------cut here---------------start------------->8---
Connecting to imap.gmail.com ([2a00:1450:400c:c00::6c]:993)... 
Connection is now encrypted
F: * OK Gimap ready for requests from 2003:df:1703:d00:5e51:4fff:fef3:ece7 b9mb127754308wmb
F: >>> 1 CAPABILITY
F: * CAPABILITY IMAP4rev1 UNSELECT IDLE NAMESPACE QUOTA ID XLIST CHILDREN X-GM-EXT-1 XYZZY SASL-IR AUTH=XOAUTH2 AUTH=PLAIN AUTH=PLAIN-CLIENTTOKEN AUTH=OAUTHBEARER
F: 1 OK Thats all she wrote! b9mb127754308wmb
Logging in...
Authenticating with SASL mechanism PLAIN...
F: >>> 2 AUTHENTICATE PLAIN <authdata>
F: * CAPABILITY IMAP4rev1 UNSELECT IDLE NAMESPACE QUOTA ID XLIST CHILDREN X-GM-EXT-1 UIDPLUS COMPRESS=DEFLATE ENABLE MOVE CONDSTORE ESEARCH UTF8=ACCEPT LIST-EXTENDED LIST-STATUS LITERAL- SPECIAL-USE APPENDLIMIT=35651584
F: 2 OK ts...@gm... authenticated (Success)
F: >>> 3 COMPRESS DEFLATE
F: 3 OK Success
F: >>> 4 ENABLE UTF8=ACCEPT
F: * ENABLED UTF8=ACCEPT
F: 4 OK Success
F: >>> 5 NAMESPACE
F: * NAMESPACE (("" "/")) NIL NIL
F: 5 OK Success
--8<---------------cut here---------------end--------------->8---

However, I don't use OAuth2 for authentication but an app password.  It
seems the plugin you've built yourself on Ubuntu is available on the
AUR:

--8<---------------cut here---------------start------------->8---
❯ yay -s xoauth2                                                                        
3 aur/sasl-xoauth2-git r203.47ff232-2 (+0 0.00) 
    SASL plugin that enables client-side use of OAuth 2.0
2 aur/oauth2ms-git r10.a1ef0cabfdea-1 (+0 0.00) (Orphaned) 
    XOAUTH2 compatible O365 token fetcher
1 aur/cyrus-sasl-xoauth2-git r24.36aabca54fd6-1 (+5 0.13) 
    XOAUTH2 mechanism plugin for cyrus-sasl
--8<---------------cut here---------------end--------------->8---

Maybe try installing that.

HTH,
  Tassilo