R: Re: R: Re: R: Re: DBERROR: critical database situation

2017-01-04 Thread absolutely_free--- via Info-cyrus 
I followed those steps:

1) shutdown imap daemon
2) rm /var/imap/db/* 
3) rm /var/imap/tls_session.db and /var/imap/deliver.db
4) reconstruct -f (as cyrus user)
5) changed /usr/local/etc/imapd.conf in this manner:

#tlscache_db: berkeley-nosync
tlscache_db: skiplist

#ptscache_db: berkeley
ptscache_db: skiplist

#duplicate_db: berkeley
duplicate_db: skiplist

6) start imap daemon

it seems quite stable.
I hope it is solved.
Thank you all for support!



Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

Re: R: Re: R: Re: DBERROR: critical database situation

2017-01-04 Thread Bron Gondwana via Info-cyrus 
Just removing the config options that say berkeley for those two databases in 
imapd.conf should do the trick (followed by nuking them again, and while Cyrus 
is shut down)

Bron.

On Wed, 4 Jan 2017, at 22:08, absolutely_free--- via Info-cyrus wrote:
> After deleting /var/imap/db/*, /var/imap/mailboxes.db and 
> /var/imap/deliver.db, 
> it rans fine for about 20 minutes, and after:
> 
> Jan  4 11:58:41 mail lmtpunix[60214]: DBERROR db5: pthread suspend failed: 
> Invalid argument
> Jan  4 11:58:41 mail lmtpunix[60214]: DBERROR db5: BDB0061 PANIC: Invalid 
> argument
> Jan  4 11:58:41 mail lmtpunix[60214]: DBERROR: critical database situation
> Jan  4 11:58:41 mail lmtpunix[60215]: DBERROR db5: BDB0060 PANIC: fatal 
> region 
> error detected; run recovery
> Jan  4 11:58:41 mail lmtpunix[60215]: DBERROR: critical database situation
> Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60215 in READY 
> state: 
> terminated abnormally
> Jan  4 11:58:41 mail lmtpunix[60217]: DBERROR db5: BDB0060 PANIC: fatal 
> region 
> error detected; run recovery
> Jan  4 11:58:41 mail lmtpunix[60217]: DBERROR: critical database situation
> Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60217 in READY 
> state: 
> terminated abnormally
> Jan  4 11:58:41 mail lmtpunix[60218]: DBERROR db5: BDB0060 PANIC: fatal 
> region 
> error detected; run recovery
> Jan  4 11:58:41 mail lmtpunix[60218]: DBERROR: critical database situation
> Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60218 in READY 
> state: 
> terminated abnormally
> Jan  4 11:58:41 mail lmtpunix[60219]: DBERROR db5: BDB0060 PANIC: fatal 
> region 
> error detected; run recovery
> Jan  4 11:58:41 mail lmtpunix[60219]: DBERROR: critical database situation
> Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60219 in READY 
> state: 
> terminated abnormally
> Jan  4 11:58:41 mail lmtpunix[60220]: DBERROR db5: BDB0060 PANIC: fatal 
> region 
> error detected; run recovery
> Jan  4 11:58:41 mail lmtpunix[60220]: DBERROR: critical database situation
> Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60220 in READY 
> state: 
> terminated abnormally
> Jan  4 11:58:41 mail lmtpunix[60221]: DBERROR db5: BDB0060 PANIC: fatal 
> region 
> error detected; run recovery
> Jan  4 11:58:41 mail lmtpunix[60221]: DBERROR: critical database situation
> Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60221 in READY 
> state: 
> terminated abnormally
> 
> 
> I am wondering what is my best option?
> Upgrade to 2.4 (at least)?
> 
> Thank you
> 
> 
> >Messaggio originale
> >Da: "Niels Dettenbach" 
> >Data: 04/01/2017 11.28
> >A: , "absolutely_f...@libero.it"
> 
> >Ogg: Re: R: Re: DBERROR: critical database situation
> >
> >Am Mittwoch, 4. Januar 2017, 11:18:42 CET schrieb absolutely_free--- via 
> Info-
> >cyrus:
> >> ./tls_sessions.db: Berkeley DB (Btree, version 9, little-endian)
> >> ./deliver.db: Berkeley DB (Btree, version 9, little-endian)
> >> 
> >> So, I can convert tls_sessions and deliver db to skiplist format, right?
> >> How can I determine which database is giving me "DBERROR: critical database
> >> situation"?
> >
> >As far as i remember, it is uncritical to just delete both files, without 
> >loosing "practical" data. tls_sessions just held details to ("running") TLS 
> >Sessions and deliver.db avoids duplicates if i remember correctly.
> >
> >Stop cyrus - do a backup of delliver.db and tls_sessions.db (or just "Move 
> it 
> >out of the way) for security (but i'm pretty shure that you can delete it) 
> or 
> >try to just delete it and restart. 
> >
> >After that try to run reconstruct -f to re-generate any indize files etc. 
> >within the mailboxes itself.
> >
> >If that not helps, come back again. ß)
> >
> >hth a bit,
> >best regards and good luck,
> >
> >Niels.
> >
> >
> >-- 
> > ---
> > Niels Dettenbach
> > Syndicat IT & Internet
> > http://www.syndicat.com
> > PGP: https://syndicat.com/pub_key.asc
> > ---
> > 
> >
> >
> >
> >
> 
> 
> 
> Cyrus Home Page: http://www.cyrusimap.org/
> List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
> To Unsubscribe:
> https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus


-- 
  Bron Gondwana
  br...@fastmail.fm

Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

Re: R: Re: DBERROR: critical database situation

2017-01-04 Thread Bron Gondwana via Info-cyrus 
Oh good, you can just delete those two BDB files with Cyrus shut down (and 
change them to skiplist while you're at it!)

Once you've done that you can delete everything in the $confdir/db folder too.

(delete might mean take a copy somewhere else until you're happy everything is 
working of course)

They're both things that don't matter across restarts.

Bron.

On Wed, 4 Jan 2017, at 21:18, absolutely_free--- via Info-cyrus wrote:
> Hi Niels,
> 
> thank you for your reply.
> I am using cyrus from sources (ports).
> 
> root@mail:/var/imap# find . -name "*.db" -exec file '{}' \;
> ./mailboxes.db: Cyrus skiplist DB
> ./annotations.db: Cyrus skiplist DB
> ./db.backup1/annotations.db: Cyrus skiplist DB
> ./db.backup1/mailboxes.db: Cyrus skiplist DB
> ./tls_sessions.db: Berkeley DB (Btree, version 9, little-endian)
> ./db.backup2/annotations.db: Cyrus skiplist DB
> ./db.backup2/mailboxes.db: Cyrus skiplist DB
> ./deliver.db: Berkeley DB (Btree, version 9, little-endian)
> 
> So, I can convert tls_sessions and deliver db to skiplist format, right?
> How can I determine which database is giving me "DBERROR: critical database 
> situation"?
> 
> Thank you
> 
> 
> Cyrus Home Page: http://www.cyrusimap.org/
> List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
> To Unsubscribe:
> https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus


-- 
  Bron Gondwana
  br...@fastmail.fm

Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

Re: R: Re: R: Re: DBERROR: critical database situation

2017-01-04 Thread Niels Dettenbach via Info-cyrus 
Am Mittwoch, 4. Januar 2017, 12:08:55 CET schrieb absolutely_free--- via Info-
cyrus:
> After deleting /var/imap/db/*, /var/imap/mailboxes.db and
> /var/imap/deliver.db, it rans fine for about 20 minutes, and after:
tls_sessions and deliver.db are not important (afaik) .

Files within ./db ARE important data files (which are only to delete when going 
the migration procedure to skiplist during that procedure). You other files 
still was skiplist (as your listing shows) - so they did not need that 
migration anymore.

Did deleting tls_sessions and deliver.db (alone and by my described procedure) 
not helped?

The error shows that there still are files within the "old" db format (i assume 
tls_sessions.db and deliver.db).

hth a bit,
good luck,


Niels.

-- 
 ---
 Niels Dettenbach
 Syndicat IT & Internet
 http://www.syndicat.com
 PGP: https://syndicat.com/pub_key.asc
 ---
 





signature.asc
Description: This is a digitally signed message part.

Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

Re: R: Re: DBERROR: critical database situation

2017-01-04 Thread Bron Gondwana via Info-cyrus 
Indeed, with all the right libraries.  You can't upgrade libdb without
everything going to shit, which is why we have ditched BDB - the upgrade
path is bogus.


Bron.





On Wed, 4 Jan 2017, at 21:20, absolutely_free--- via Info-cyrus wrote:

> 

> Hi Bron,

> thank you for you reply.

> What do you mean with "get some version 10 binaries"?

> Do you mean "binaries from FreeBSD 10"?

> 
> Thank you

> 

> 

> Cyrus Home Page: http://www.cyrusimap.org/

> List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
> To Unsubscribe:

> https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus



--

  Bron Gondwana

  br...@fastmail.fm





Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

R: Re: R: Re: DBERROR: critical database situation

2017-01-04 Thread absolutely_free--- via Info-cyrus 
After deleting /var/imap/db/*, /var/imap/mailboxes.db and /var/imap/deliver.db, 
it rans fine for about 20 minutes, and after:

Jan  4 11:58:41 mail lmtpunix[60214]: DBERROR db5: pthread suspend failed: 
Invalid argument
Jan  4 11:58:41 mail lmtpunix[60214]: DBERROR db5: BDB0061 PANIC: Invalid 
argument
Jan  4 11:58:41 mail lmtpunix[60214]: DBERROR: critical database situation
Jan  4 11:58:41 mail lmtpunix[60215]: DBERROR db5: BDB0060 PANIC: fatal region 
error detected; run recovery
Jan  4 11:58:41 mail lmtpunix[60215]: DBERROR: critical database situation
Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60215 in READY state: 
terminated abnormally
Jan  4 11:58:41 mail lmtpunix[60217]: DBERROR db5: BDB0060 PANIC: fatal region 
error detected; run recovery
Jan  4 11:58:41 mail lmtpunix[60217]: DBERROR: critical database situation
Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60217 in READY state: 
terminated abnormally
Jan  4 11:58:41 mail lmtpunix[60218]: DBERROR db5: BDB0060 PANIC: fatal region 
error detected; run recovery
Jan  4 11:58:41 mail lmtpunix[60218]: DBERROR: critical database situation
Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60218 in READY state: 
terminated abnormally
Jan  4 11:58:41 mail lmtpunix[60219]: DBERROR db5: BDB0060 PANIC: fatal region 
error detected; run recovery
Jan  4 11:58:41 mail lmtpunix[60219]: DBERROR: critical database situation
Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60219 in READY state: 
terminated abnormally
Jan  4 11:58:41 mail lmtpunix[60220]: DBERROR db5: BDB0060 PANIC: fatal region 
error detected; run recovery
Jan  4 11:58:41 mail lmtpunix[60220]: DBERROR: critical database situation
Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60220 in READY state: 
terminated abnormally
Jan  4 11:58:41 mail lmtpunix[60221]: DBERROR db5: BDB0060 PANIC: fatal region 
error detected; run recovery
Jan  4 11:58:41 mail lmtpunix[60221]: DBERROR: critical database situation
Jan  4 11:58:41 mail master[60046]: service lmtpunix pid 60221 in READY state: 
terminated abnormally


I am wondering what is my best option?
Upgrade to 2.4 (at least)?

Thank you


>Messaggio originale
>Da: "Niels Dettenbach" 
>Data: 04/01/2017 11.28
>A: , "absolutely_f...@libero.it"

>Ogg: Re: R: Re: DBERROR: critical database situation
>
>Am Mittwoch, 4. Januar 2017, 11:18:42 CET schrieb absolutely_free--- via 
Info-
>cyrus:
>> ./tls_sessions.db: Berkeley DB (Btree, version 9, little-endian)
>> ./deliver.db: Berkeley DB (Btree, version 9, little-endian)
>> 
>> So, I can convert tls_sessions and deliver db to skiplist format, right?
>> How can I determine which database is giving me "DBERROR: critical database
>> situation"?
>
>As far as i remember, it is uncritical to just delete both files, without 
>loosing "practical" data. tls_sessions just held details to ("running") TLS 
>Sessions and deliver.db avoids duplicates if i remember correctly.
>
>Stop cyrus - do a backup of delliver.db and tls_sessions.db (or just "Move 
it 
>out of the way) for security (but i'm pretty shure that you can delete it) 
or 
>try to just delete it and restart. 
>
>After that try to run reconstruct -f to re-generate any indize files etc. 
>within the mailboxes itself.
>
>If that not helps, come back again. ß)
>
>hth a bit,
>best regards and good luck,
>
>Niels.
>
>
>-- 
> ---
> Niels Dettenbach
> Syndicat IT & Internet
> http://www.syndicat.com
> PGP: https://syndicat.com/pub_key.asc
> ---
> 
>
>
>
>



Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

Re: R: Re: DBERROR: critical database situation

2017-01-04 Thread Niels Dettenbach via Info-cyrus 
Am Mittwoch, 4. Januar 2017, 11:18:42 CET schrieb absolutely_free--- via Info-
cyrus:
> ./tls_sessions.db: Berkeley DB (Btree, version 9, little-endian)
> ./deliver.db: Berkeley DB (Btree, version 9, little-endian)
> 
> So, I can convert tls_sessions and deliver db to skiplist format, right?
> How can I determine which database is giving me "DBERROR: critical database
> situation"?

As far as i remember, it is uncritical to just delete both files, without 
loosing "practical" data. tls_sessions just held details to ("running") TLS 
Sessions and deliver.db avoids duplicates if i remember correctly.

Stop cyrus - do a backup of delliver.db and tls_sessions.db (or just "Move it 
out of the way) for security (but i'm pretty shure that you can delete it) or 
try to just delete it and restart. 

After that try to run reconstruct -f to re-generate any indize files etc. 
within the mailboxes itself.

If that not helps, come back again. ß)

hth a bit,
best regards and good luck,

Niels.


-- 
 ---
 Niels Dettenbach
 Syndicat IT & Internet
 http://www.syndicat.com
 PGP: https://syndicat.com/pub_key.asc
 ---
 





signature.asc
Description: This is a digitally signed message part.

Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

R: Re: DBERROR: critical database situation

2017-01-04 Thread absolutely_free--- via Info-cyrus 
Hi Bron,thank you for you reply.What do you mean with "get some version 10 
binaries"?Do you mean "binaries from FreeBSD 10"?
Thank you


Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

R: Re: DBERROR: critical database situation

2017-01-04 Thread absolutely_free--- via Info-cyrus 
Hi Niels,

thank you for your reply.
I am using cyrus from sources (ports).

root@mail:/var/imap# find . -name "*.db" -exec file '{}' \;
./mailboxes.db: Cyrus skiplist DB
./annotations.db: Cyrus skiplist DB
./db.backup1/annotations.db: Cyrus skiplist DB
./db.backup1/mailboxes.db: Cyrus skiplist DB
./tls_sessions.db: Berkeley DB (Btree, version 9, little-endian)
./db.backup2/annotations.db: Cyrus skiplist DB
./db.backup2/mailboxes.db: Cyrus skiplist DB
./deliver.db: Berkeley DB (Btree, version 9, little-endian)

So, I can convert tls_sessions and deliver db to skiplist format, right?
How can I determine which database is giving me "DBERROR: critical database 
situation"?

Thank you


Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

Re: DBERROR: critical database situation

2017-01-04 Thread Bron Gondwana via Info-cyrus 
2.4 was releasted in 2010, it's 2017.  We advise not to run Berkeley DB
  (even on 2.3.19), because it doesn't upgrade very nicely.


I would advise that you get some version 10 binaries and use cvt_cyrusdb
to convert all your berkeley databases to skiplist. Skiplist in 2.3.19
is rock solid.  You'll need to update the config as well, but the
default for every database type in 2.3.19 is skiplist, so it's just a
matter of removing some lines from imapd.conf.


Regards,



Bron.



On Wed, 4 Jan 2017, at 19:45, absolutely_free--- via Info-cyrus wrote:

> 

> Hi,

> I am using cyrus-imapd23-2.3.19_2 on FreeBSD.

> After BSD upgrade (from 10 to 11) I get problems with cyrus.

> I get this kind of errors on:

> 

> Jan  4 09:27:31 mail imaps[65141]: DBERROR db5: pthread suspend
> failed: Invalid argument
> Jan  4 09:27:31 mail imaps[65141]: DBERROR db5: BDB0061 PANIC: Invalid
> argument
> Jan  4 09:27:31 mail imaps[65141]: DBERROR: critical database
> situation
> Jan  4 09:27:31 mail imaps[65140]: DBERROR db5: BDB0060 PANIC: fatal
> region error detected; run recovery
> Jan  4 09:27:31 mail imaps[65140]: DBERROR: critical database
> situation
> Jan  4 09:27:38 mail imaps[65142]: DBERROR db5: BDB0060 PANIC: fatal
> region error detected; run recovery
> Jan  4 09:27:38 mail imaps[65142]: DBERROR: critical database
> situation
> Jan  4 09:27:38 mail master[65081]: service imaps pid 65142 in READY
> state: terminated abnormally
> Jan  4 09:27:38 mail imaps[65143]: DBERROR db5: BDB0060 PANIC: fatal
> region error detected; run recovery
> Jan  4 09:27:38 mail imaps[65143]: DBERROR: critical database
> situation
> Jan  4 09:27:38 mail master[65081]: service imaps pid 65143 in READY
> state: terminated abnormally
> 

> So I stopped imapd service, and ran /usr/local/cyrus/bin/ctl_cyrusdb
> -r as cyrus user.
> 

> Rebuild went fine: 

> 

> Jan  4 09:36:17 mail ctl_cyrusdb[62332]: recovering cyrus databases

> Jan  4 09:36:17 mail ctl_cyrusdb[62332]: done recovering cyrus
> databases
> 

> and I also checked that just after ctl_cyrusdb command, I was able to
> dump mailboxes.db content with ctl_mboxlist -d
> 

> After few minutes, I get same errors, and ctl_mboxlist -d returns
> nothing:
> 

> cyrus@mail:/root$ /usr/local/cyrus/bin/ctl_cyrusdb -r

> cyrus@mail:/root$ 

> 

> Yesterday I tried to:

> 

> stop services

> remove and recreate /var/imap folder

> run reconstruct

> 

> but it seems I got no stable solution

> 

> Can you help me to solve this problem?

> Thank you very much

> 

> This is my imapd.conf:

> 

> root@mail:/usr/local/etc# grep -v ^# imapd.conf |grep -v ^$

> configdirectory: /var/imap

> partition-default: /var/spool/imap

> allowapop: 0

> unixhierarchysep: no

> altnamespace: no

> allowanonymouslogin: no

> allowplaintext: yes

> quotawarn: 90

> timeout: 30

> imapidlepoll: 60

> poptimeout: 10

> popminpoll: 0

> admins: root cyrus

> defaultacl: anyone lrs

> duplicatesuppression: no

> sievedir: /var/imap/sieve

> postmaster: postmaster

> sieve_maxscriptsize: 32

> sieve_maxscripts: 5

> duplicate_db: berkeley

> mboxlist_db: skiplist

> ptscache_db: berkeley

> seenstate_db: skiplist

> sasl_pwcheck_method: saslauthd

> sasl_mech_list: plain

> tls_cert_file: /etc/certs/ssl.crt/server.crt

> tls_key_file: /etc/certs/ssl.key/server.key

> tls_ca_file:  /etc/certs/ssl.crt/gd_bundle-g2-g1.crt

> tls_ca_path: /etc/certs/ssl.crt

> notifysocket: /var/imap/socket/notify

> 

> And cyrus.conf

> 

> root@mail:/usr/local/etc# grep -v ^# cyrus.conf |grep -v ^$
> START {

>   # do not delete this entry!

>   recover   cmd="ctl_cyrusdb -r"

>   # this is only necessary if using idled for IMAP IDLE

> }

> SERVICES {

>   # add or remove based on preferences

>   imap  cmd="imapd" listen="127.0.0.1:imap" prefork=1

>   imaps cmd="imapd -s" listen="imaps" prefork=0

>   #pop3 cmd="pop3d" listen="pop3" prefork=0

>   pop3s cmd="pop3d -s" listen="pop3s" prefork=0

>   sieve cmd="timsieved" listen="sieve" prefork=0

>   # these are only necessary if receiving/exporting usenet via NNTP

>   # at least one LMTP is required for delivery

>   lmtpunix  cmd="lmtpd" listen="/var/imap/socket/lmtp" prefork=0

>   # this is required if using notifications

> }

> EVENTS {

>   # this is required

>   checkpointcmd="ctl_cyrusdb -c" period=30

>   # this is only necessary if using duplicate delivery suppression,

>   # Sieve or NNTP

>   delprune  cmd="cyr_expire -E 3" at=0400

>   # this is only necessary if caching TLS sessions

>   tlsprune  cmd="tls_prune" at=0400

> }

> 

> 

> 

> 

> 

> 

> Cyrus Home Page: http://www.cyrusimap.org/

> List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
> To Unsubscribe:

> https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus



--

  Bron Gondwana

  br...@fastmail.fm





Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info:

Re: DBERROR: critical database situation

2017-01-04 Thread Niels Dettenbach via Info-cyrus 
Am Mittwoch, 4. Januar 2017, 09:45:47 CET schrieb absolutely_free--- via Info-
cyrus:
> Hi,
> I am using cyrus-imapd23-2.3.19_2 on FreeBSD.
> After BSD upgrade (from 10 to 11) I get problems with cyrus.
> I get this kind of errors on:
> 
> Jan  4 09:27:31 mail imaps[65141]: DBERROR db5: pthread suspend failed:
> Invalid argument Jan  4 09:27:31 mail imaps[65141]: DBERROR db5: BDB0061
> PANIC: Invalid argument Jan  4 09:27:31 mail imaps[65141]: DBERROR:
> critical database situation Jan  4 09:27:31 mail imaps[65140]: DBERROR db5:
> BDB0060 PANIC: fatal region error detected; run recovery Jan  4 09:27:31
> mail imaps[65140]: DBERROR: critical database situation Jan  4 09:27:38
> mail imaps[65142]: DBERROR db5: BDB0060 PANIC: fatal region error detected;
> run recovery Jan  4 09:27:38 mail imaps[65142]: DBERROR: critical database
> situation Jan  4 09:27:38 mail master[65081]: service imaps pid 65142 in
> READY state: terminated abnormally Jan  4 09:27:38 mail imaps[65143]:
> DBERROR db5: BDB0060 PANIC: fatal region error detected; run recovery Jan 
> 4 09:27:38 mail imaps[65143]: DBERROR: critical database situation Jan  4
> 09:27:38 mail master[65081]: service imaps pid 65143 in READY state:
> terminated abnormally

Just to clearify here: it is important to follow the upgrade procedure from 
cyrus:
https://www.cyrusimap.org/docs/cyrus-imapd/2.4.0/install-upgrade.php

and to export/reimport any data during db version changes / upgrades within 
the used cyrus builds (export before - delete db and and reimport after).

see i.e:
https://lists.andrew.cmu.edu/pipermail/info-cyrus/2015-August/038316.html
https://www.banquise.org/software/how-to-recover-from-cyrus-when-you-have-some-db-errors/
 (Linux, but same thing...)


Another thing often required in such a situation is to remove :

rm -f /var/lib/cyrus/deliver.db
rm -f /var/lib/cyrus/tls_sessions.db

to avoid the existence of incompatible database files.

If that doesn't help, a bug in your build may be a reason. Did you build the 
software from sources / ports or installed binaries?

hth a bit,

Niels.

-- 
 ---
 Niels Dettenbach
 Syndicat IT & Internet
 http://www.syndicat.com
 PGP: https://syndicat.com/pub_key.asc
 ---
 





signature.asc
Description: This is a digitally signed message part.

Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus

DBERROR: critical database situation

2017-01-04 Thread absolutely_free--- via Info-cyrus 
Hi,I am using cyrus-imapd23-2.3.19_2 on FreeBSD.After BSD upgrade (from 10 to 
11) I get problems with cyrus.I get this kind of errors on:
Jan  4 09:27:31 mail imaps[65141]: DBERROR db5: pthread suspend failed: Invalid 
argumentJan  4 09:27:31 mail imaps[65141]: DBERROR db5: BDB0061 PANIC: Invalid 
argumentJan  4 09:27:31 mail imaps[65141]: DBERROR: critical database 
situationJan  4 09:27:31 mail imaps[65140]: DBERROR db5: BDB0060 PANIC: fatal 
region error detected; run recoveryJan  4 09:27:31 mail imaps[65140]: DBERROR: 
critical database situationJan  4 09:27:38 mail imaps[65142]: DBERROR db5: 
BDB0060 PANIC: fatal region error detected; run recoveryJan  4 09:27:38 mail 
imaps[65142]: DBERROR: critical database situationJan  4 09:27:38 mail 
master[65081]: service imaps pid 65142 in READY state: terminated abnormallyJan 
 4 09:27:38 mail imaps[65143]: DBERROR db5: BDB0060 PANIC: fatal region error 
detected; run recoveryJan  4 09:27:38 mail imaps[65143]: DBERROR: critical 
database situationJan  4 09:27:38 mail master[65081]: service imaps pid 65143 
in READY state: terminated abnormally
So I stopped imapd service, and ran /usr/local/cyrus/bin/ctl_cyrusdb -r as 
cyrus user.
Rebuild went fine: 
Jan  4 09:36:17 mail ctl_cyrusdb[62332]: recovering cyrus databasesJan  4 
09:36:17 mail ctl_cyrusdb[62332]: done recovering cyrus databases
and I also checked that just after ctl_cyrusdb command, I was able to dump 
mailboxes.db content with ctl_mboxlist -d
After few minutes, I get same errors, and ctl_mboxlist -d returns nothing:
cyrus@mail:/root$ /usr/local/cyrus/bin/ctl_cyrusdb -rcyrus@mail:/root$ 
Yesterday I tried to:
stop servicesremove and recreate /var/imap folderrun reconstruct
but it seems I got no stable solution
Can you help me to solve this problem?Thank you very much
This is my imapd.conf:
root@mail:/usr/local/etc# grep -v ^# imapd.conf |grep -v ^$configdirectory: 
/var/imappartition-default: /var/spool/imapallowapop: 0unixhierarchysep: 
noaltnamespace: noallowanonymouslogin: noallowplaintext: yesquotawarn: 
90timeout: 30imapidlepoll: 60poptimeout: 10popminpoll: 0admins: root 
cyrusdefaultacl: anyone lrsduplicatesuppression: nosievedir: 
/var/imap/sievepostmaster: postmastersieve_maxscriptsize: 32sieve_maxscripts: 
5duplicate_db: berkeleymboxlist_db: skiplistptscache_db: berkeleyseenstate_db: 
skiplistsasl_pwcheck_method: saslauthdsasl_mech_list: plaintls_cert_file: 
/etc/certs/ssl.crt/server.crttls_key_file: 
/etc/certs/ssl.key/server.keytls_ca_file:  
/etc/certs/ssl.crt/gd_bundle-g2-g1.crttls_ca_path: 
/etc/certs/ssl.crtnotifysocket: /var/imap/socket/notify
And cyrus.conf
root@mail:/usr/local/etc# grep -v ^# cyrus.conf |grep -v ^$  START {  # 
do not delete this entry!  recover   cmd="ctl_cyrusdb -r"  # this is only 
necessary if using idled for IMAP IDLE}SERVICES {  # add or remove based on 
preferences  imap  cmd="imapd" listen="127.0.0.1:imap" prefork=1  imaps 
cmd="imapd -s" listen="imaps" prefork=0  #pop3 cmd="pop3d" 
listen="pop3" prefork=0  pop3s cmd="pop3d -s" listen="pop3s" prefork=0  
sieve cmd="timsieved" listen="sieve" prefork=0  # these are only 
necessary if receiving/exporting usenet via NNTP  # at least one LMTP is 
required for delivery  lmtpunix  cmd="lmtpd" listen="/var/imap/socket/lmtp" 
prefork=0  # this is required if using notifications}EVENTS {  # this is 
required  checkpointcmd="ctl_cyrusdb -c" period=30  # this is only 
necessary if using duplicate delivery suppression,  # Sieve or NNTP  delprune   
   cmd="cyr_expire -E 3" at=0400  # this is only necessary if caching TLS 
sessions  tlsprune  cmd="tls_prune" at=0400}





Cyrus Home Page: http://www.cyrusimap.org/
List Archives/Info: http://lists.andrew.cmu.edu/pipermail/info-cyrus/
To Unsubscribe:
https://lists.andrew.cmu.edu/mailman/listinfo/info-cyrus