Discussion:
Guam broken in Centos 7
Nik Middleton
2017-07-07 21:24:02 UTC
Permalink
Guam is broken in Centos 7 installs so if you’ve done a yum update and your system is trashed, you need to downgrade Guam to 0.8.3



http://linuxdoku.bplaced.net/?p=401 has the rpm. Remove Guam without deleting dependencies and it will work again



Regards,
Martin Campbell
2017-07-08 21:54:05 UTC
Permalink
I'm using Kolab 16 on CentOS 7.

Since upgrading to guam-0.9.2-2.1.el7.kolab_16.x86_64, I've had terrible
IMAP performance.

I presume that's the same problem.
Post by Nik Middleton
Guam is broken in Centos 7 installs so if you’ve done a yum update and
your system is trashed, you need to downgrade Guam to 0.8.3
https://u2619215.ct.sendgrid.net/wf/click?upn=xyHy6Gye8FSpcIWrVzOzlLWdPGKSw9kKJzMXXa3y9xHy1bRzAQd0AotOyHE2efks_oEMJ28Tbv3NCKzw6SHcfLsfMId7-2BGeFMlh-2Bx90-2F-2FaWvxIlFDXeH6nXHnUPV4faJ5MX1OmGHNp3tP82-2FUDCWaKog2CMLvnaqDXrmucXEShwpzJ1bbj6Fjc9eEQSNmtFvSFxRAMXQXa-2BLxO2OaOouzqh8JmcJ99hwCKvfIUNbFZbpkEW9DKA4C5oIcxZTCrPq0gj52EjAA0Kvj57ISddlCMR-2BvUdC4Vnzq-2FLsZyLDrMAc-3D has the rpm. Remove Guam without
deleting dependencies and it will work again
Nik Middleton
2017-07-08 22:30:53 UTC
Permalink
I would think so.  Guam will crash and then restart so lots of spinning wheels on user’s screens.  It wasn’t just the slow performance for us.  We have a mix of systems and clients.  Macs were Ok with Outlook, but Windows PC’s (7&10) would not work at all with Outlook.  Thunderbird was fine on Windows.



The error messages were misleading (bad CA etc.) and it took two days to figure out what part of the system was at fault.



Regards





From: <***@placefarm.com> on behalf of Martin Campbell <***@placefarm.com>
Date: Saturday, 8 July 2017 at 22:54
To: Nik Middleton <***@antheus.co.uk>, <***@lists.kolab.org>
Subject: Re: Guam broken in Centos 7



I'm using Kolab 16 on CentOS 7.

Since upgrading to guam-0.9.2-2.1.el7.kolab_16.x86_64, I've had terrible IMAP performance.

I presume that's the same problem.



On 07/07/2017 22:24, Nik Middleton wrote:

Guam is broken in Centos 7 installs so if you’ve done a yum update and your system is trashed, you need to downgrade Guam to 0.8.3



http://linuxdoku.bplaced.net/?p=401 has the rpm. Remove Guam without deleting dependencies and it will work again
Roland Kolb
2017-07-09 16:51:15 UTC
Permalink
Hi Nik,

due to we have also this performance issue (up to 15min to get mails)
since the upgrade to guam-0.9.2 I think also to owngrade Guam to 0.8.3
Changes in kolab.conf don't help as it helped during the last
performance issue with guam one year ago.
What is the best way for downgrading guam you suggest?

Thanks

Roland
I would think so. Guam will crash and then restart so lots of
spinning wheels on user’s screens. It wasn’t just the slow
performance for us. We have a mix of systems and clients. Macs were
Ok with Outlook, but Windows PC’s (7&10) would not work at all with
Outlook. Thunderbird was fine on Windows.
The error messages were misleading (bad CA etc.) and it took two days
to figure out what part of the system was at fault.
Regards
*Date: *Saturday, 8 July 2017 at 22:54
*Subject: *Re: Guam broken in Centos 7
I'm using Kolab 16 on CentOS 7.
Since upgrading to guam-0.9.2-2.1.el7.kolab_16.x86_64, I've had terrible IMAP performance.
I presume that's the same problem.
Guam is broken in Centos 7 installs so if you’ve done a yum update
and your system is trashed, you need to downgrade Guam to 0.8.3
http://linuxdoku.bplaced.net/?p=401
<https://u2619215.ct.sendgrid.net/wf/click?upn=xyHy6Gye8FSpcIWrVzOzlLWdPGKSw9kKJzMXXa3y9xHy1bRzAQd0AotOyHE2efks_PUFsV5AZSt-2FEXoAPnk9gfI38782PLGBQV9UCRgwrAWVlbDyDP9SARQ0S4-2BV2GQIrvaT9vMpuGFe9Vs6WQmQcX81YvlIpqsqshUErRzhcVy7ZI8FKof1aF7aUo0P5bgYs1nCUpP-2BczbKoAd1PwpYA-2Bb7M18xfBswGM7P2xARpCFPEEcuHAZ-2FJJ8KUS2egmk-2B0muhaBQurcPnChRivllh87Hb8f-2BxuH4uQLQKiTorYAQU-3D>
has the rpm. Remove Guam without deleting dependencies and it
will work again
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
Brady, Mike
2017-07-09 19:18:18 UTC
Permalink
Download the 0.8.3 rpm if you don't have it locally and do

yum downgrade guam-0.8.3-2.1.el7.centos.kolab_16.x86_64.rpm

I also blocked updates to guam for now with versionlock

yum install yum-plugin-versionlock.noarch
yum versionlock add guam
yum versionlock list

guam 0.9.2 is just plain broken. There is also this post in the forum
https://kolab.org/hub/topic/264/guam-error-after-update-to-0-9-2

Also the KolabNow service only recently implemented Kolab16 including
guam (don't know what release) and they had to disable it after only a
few days of use
(https://kanarip.wordpress.com/2017/07/03/kolab-now-disabled-imap-proxy/).


Winterfell only has guam 0.9.0 in it. This puzzles me as I thought that
Winterfell was supposed to be the development repo, so I do not
understand how/why the Kolab16 repo would have a later version.
Winterfell does have a later version of erlang, so maybe there is a
version incompatibility?

We have also had to disable Wallace at it constantly (several times a
week on a 50 user system) stopped processing emails.
Post by Roland Kolb
Hi Nik,
due to we have also this performance issue (up to 15min to get mails) since the upgrade to guam-0.9.2 I think also to owngrade Guam to 0.8.3
Changes in kolab.conf don't help as it helped during the last performance issue with guam one year ago.
What is the best way for downgrading guam you suggest?
Thanks
Roland
I would think so. Guam will crash and then restart so lots of spinning wheels on user's screens. It wasn't just the slow performance for us. We have a mix of systems and clients. Macs were Ok with Outlook, but Windows PC's (7&10) would not work at all with Outlook. Thunderbird was fine on Windows.
The error messages were misleading (bad CA etc.) and it took two days to figure out what part of the system was at fault.
Regards
DATE: Saturday, 8 July 2017 at 22:54
SUBJECT: Re: Guam broken in Centos 7
I'm using Kolab 16 on CentOS 7.
Since upgrading to guam-0.9.2-2.1.el7.kolab_16.x86_64, I've had terrible IMAP performance.
I presume that's the same problem.
Guam is broken in Centos 7 installs so if you've done a yum update and your system is trashed, you need to downgrade Guam to 0.8.3
http://linuxdoku.bplaced.net/?p=401 [1] has the rpm. Remove Guam without deleting dependencies and it will work again
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
_______________________________________________
users mailing list
***@lists.kolab.org
https://lists.kolab.org/mailman/listinfo/users



Links:
------
[1]
https://u2619215.ct.sendgrid.net/wf/click?upn=xyHy6Gye8FSpcIWrVzOzlLWdPGKSw9kKJzMXXa3y9xHy1bRzAQd0AotOyHE2efks_PUFsV5AZSt-2FEXoAPnk9gfI38782PLGBQV9UCRgwrAWVlbDyDP9SARQ0S4-2BV2GQIrvaT9vMpuGFe9Vs6WQmQcX81YvlIpqsqshUErRzhcVy7ZI8FKof1aF7aUo0P5bgYs1nCUpP-2BczbKoAd1PwpYA-2Bb7M18xfBswGM7P2xARpCFPEEcuHAZ-2FJJ8KUS2egmk-2B0muhaBQurcPnChRivllh87Hb8f-2BxuH4uQLQKiTorYAQU-3D
Skale, Franz
2017-07-10 06:27:40 UTC
Permalink
Hi,
we had already a thread regarding guam.
After the update 0.9.2-1 to 0.9.2.2 the errors were gone but guam
stopped working as it endless looped in a threat never meant to join.
I cannot debug erlang but a simple strace shows the problem.
I don't know why erlang is used to do proxying and it seems that the
programmers do have real troubles to get it working. (Thunderbird issues
etc.).
Disabling guam and halting the package works for me though.
I cannot copy your problem with wallace so far i have to admit. (100
Users).
I use debian jessie.

Rgds.
Franz
Post by Brady, Mike
Download the 0.8.3 rpm if you don't have it locally and do
yum downgrade guam-0.8.3-2.1.el7.centos.kolab_16.x86_64.rpm
I also blocked updates to guam for now with versionlock
yum install yum-plugin-versionlock.noarch
yum versionlock add guam
yum versionlock list
guam 0.9.2 is just plain broken. There is also this post in the forum
https://kolab.org/hub/topic/264/guam-error-after-update-to-0-9-2
Also the KolabNow service only recently implemented Kolab16 including
guam (don't know what release) and they had to disable it after only a
few days of use
(https://kanarip.wordpress.com/2017/07/03/kolab-now-disabled-imap-proxy/).
Winterfell only has guam 0.9.0 in it. This puzzles me as I thought
that Winterfell was supposed to be the development repo, so I do not
understand how/why the Kolab16 repo would have a later version.
Winterfell does have a later version of erlang, so maybe there is a
version incompatibility?
We have also had to disable Wallace at it constantly (several times a
week on a 50 user system) stopped processing emails.
Post by Roland Kolb
Hi Nik,
due to we have also this performance issue (up to 15min to get
mails) since the upgrade to guam-0.9.2 I think also to owngrade Guam
to 0.8.3
Changes in kolab.conf don't help as it helped during the last
performance issue with guam one year ago.
What is the best way for downgrading guam you suggest?
Thanks
Roland
I would think so. Guam will crash and then restart so lots of
spinning wheels on user's screens. It wasn't just the slow
performance for us. We have a mix of systems and clients. Macs
were Ok with Outlook, but Windows PC's (7&10) would not work at all
with Outlook. Thunderbird was fine on Windows.
The error messages were misleading (bad CA etc.) and it took two
days to figure out what part of the system was at fault.
Regards
DATE: Saturday, 8 July 2017 at 22:54
SUBJECT: Re: Guam broken in Centos 7
I'm using Kolab 16 on CentOS 7.
Since upgrading to guam-0.9.2-2.1.el7.kolab_16.x86_64, I've had
terrible IMAP performance.
I presume that's the same problem.
Guam is broken in Centos 7 installs so if you've done a yum update
and your system is trashed, you need to downgrade Guam to 0.8.3
http://linuxdoku.bplaced.net/?p=401 [1] has the rpm. Remove Guam
without deleting dependencies and it will work again
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
------
[1]
https://u2619215.ct.sendgrid.net/wf/click?upn=xyHy6Gye8FSpcIWrVzOzlLWdPGKSw9kKJzMXXa3y9xHy1bRzAQd0AotOyHE2efks_PUFsV5AZSt-2FEXoAPnk9gfI38782PLGBQV9UCRgwrAWVlbDyDP9SARQ0S4-2BV2GQIrvaT9vMpuGFe9Vs6WQmQcX81YvlIpqsqshUErRzhcVy7ZI8FKof1aF7aUo0P5bgYs1nCUpP-2BczbKoAd1PwpYA-2Bb7M18xfBswGM7P2xARpCFPEEcuHAZ-2FJJ8KUS2egmk-2B0muhaBQurcPnChRivllh87Hb8f-2BxuH4uQLQKiTorYAQU-3D
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
Nik Middleton
2017-07-10 10:31:14 UTC
Permalink
Hi,

When you say you disable Guam, are you simply stopping start on boot up and changing the Cyrus ports?

I must admit, this issue with Guam has a very low visibility it took me sometime to figure out what the problem was. This issue should flagged on the main site.

Regards


On 10/07/2017, 07:27, "Skale, Franz" <users-***@lists.kolab.org on behalf of ***@dah.am> wrote:

Hi,
we had already a thread regarding guam.
After the update 0.9.2-1 to 0.9.2.2 the errors were gone but guam
stopped working as it endless looped in a threat never meant to join.
I cannot debug erlang but a simple strace shows the problem.
I don't know why erlang is used to do proxying and it seems that the
programmers do have real troubles to get it working. (Thunderbird issues
etc.).
Disabling guam and halting the package works for me though.
I cannot copy your problem with wallace so far i have to admit. (100
Users).
I use debian jessie.

Rgds.
Franz
Post by Brady, Mike
Download the 0.8.3 rpm if you don't have it locally and do
yum downgrade guam-0.8.3-2.1.el7.centos.kolab_16.x86_64.rpm
I also blocked updates to guam for now with versionlock
yum install yum-plugin-versionlock.noarch
yum versionlock add guam
yum versionlock list
guam 0.9.2 is just plain broken. There is also this post in the forum
https://kolab.org/hub/topic/264/guam-error-after-update-to-0-9-2
Also the KolabNow service only recently implemented Kolab16 including
guam (don't know what release) and they had to disable it after only a
few days of use
(https://kanarip.wordpress.com/2017/07/03/kolab-now-disabled-imap-proxy/).
Winterfell only has guam 0.9.0 in it. This puzzles me as I thought
that Winterfell was supposed to be the development repo, so I do not
understand how/why the Kolab16 repo would have a later version.
Winterfell does have a later version of erlang, so maybe there is a
version incompatibility?
We have also had to disable Wallace at it constantly (several times a
week on a 50 user system) stopped processing emails.
Post by Roland Kolb
Hi Nik,
due to we have also this performance issue (up to 15min to get
mails) since the upgrade to guam-0.9.2 I think also to owngrade Guam
to 0.8.3
Changes in kolab.conf don't help as it helped during the last
performance issue with guam one year ago.
What is the best way for downgrading guam you suggest?
Thanks
Roland
I would think so. Guam will crash and then restart so lots of
spinning wheels on user's screens. It wasn't just the slow
performance for us. We have a mix of systems and clients. Macs
were Ok with Outlook, but Windows PC's (7&10) would not work at all
with Outlook. Thunderbird was fine on Windows.
The error messages were misleading (bad CA etc.) and it took two
days to figure out what part of the system was at fault.
Regards
DATE: Saturday, 8 July 2017 at 22:54
SUBJECT: Re: Guam broken in Centos 7
I'm using Kolab 16 on CentOS 7.
Since upgrading to guam-0.9.2-2.1.el7.kolab_16.x86_64, I've had
terrible IMAP performance.
I presume that's the same problem.
Guam is broken in Centos 7 installs so if you've done a yum update
and your system is trashed, you need to downgrade Guam to 0.8.3
http://linuxdoku.bplaced.net/?p=401 [1] has the rpm. Remove Guam
without deleting dependencies and it will work again
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
------
[1]
https://u2619215.ct.sendgrid.net/wf/click?upn=xyHy6Gye8FSpcIWrVzOzlLWdPGKSw9kKJzMXXa3y9xHy1bRzAQd0AotOyHE2efks_PUFsV5AZSt-2FEXoAPnk9gfI38782PLGBQV9UCRgwrAWVlbDyDP9SARQ0S4-2BV2GQIrvaT9vMpuGFe9Vs6WQmQcX81YvlIpqsqshUErRzhcVy7ZI8FKof1aF7aUo0P5bgYs1nCUpP-2BczbKoAd1PwpYA-2Bb7M18xfBswGM7P2xARpCFPEEcuHAZ-2FJJ8KUS2egmk-2B0muhaBQurcPnChRivllh87Hb8f-2BxuH4uQLQKiTorYAQU-3D
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
_______________________________________________
users mailing list
***@lists.kolab.org
https://lists.kolab.org/mailman/listinfo/users
Skale, Franz
2017-07-10 12:21:10 UTC
Permalink
Hi Nik,
yes, do:
systemctl disable guam and fix the ports in cyrus.conf.
There's really very low programmer output.
When you finish a software or a part you do extensive testing which
fails immediately when using a simple mail client.
Don't know why it has to be erlang which is hard to debug other than
unjoined threats.
Also the git is not very meaningful.
Last checkin 3 may.
See:
https://git.kolab.org/diffusion/G/

Rgds.

Franz
Post by Nik Middleton
Hi,
When you say you disable Guam, are you simply stopping start on boot
up and changing the Cyrus ports?
I must admit, this issue with Guam has a very low visibility it took
me sometime to figure out what the problem was. This issue should
flagged on the main site.
Regards
Hi,
we had already a thread regarding guam.
After the update 0.9.2-1 to 0.9.2.2 the errors were gone but guam
stopped working as it endless looped in a threat never meant to join.
I cannot debug erlang but a simple strace shows the problem.
I don't know why erlang is used to do proxying and it seems that the
programmers do have real troubles to get it working. (Thunderbird issues
etc.).
Disabling guam and halting the package works for me though.
I cannot copy your problem with wallace so far i have to admit. (100
Users).
I use debian jessie.
Rgds.
Franz
Post by Brady, Mike
Download the 0.8.3 rpm if you don't have it locally and do
yum downgrade guam-0.8.3-2.1.el7.centos.kolab_16.x86_64.rpm
I also blocked updates to guam for now with versionlock
yum install yum-plugin-versionlock.noarch
yum versionlock add guam
yum versionlock list
guam 0.9.2 is just plain broken. There is also this post in the
forum
Post by Brady, Mike
https://kolab.org/hub/topic/264/guam-error-after-update-to-0-9-2
Also the KolabNow service only recently implemented Kolab16
including
Post by Brady, Mike
guam (don't know what release) and they had to disable it after
only a
Post by Brady, Mike
few days of use
(https://kanarip.wordpress.com/2017/07/03/kolab-now-disabled-imap-proxy/).
Post by Brady, Mike
Winterfell only has guam 0.9.0 in it. This puzzles me as I
thought
Post by Brady, Mike
that Winterfell was supposed to be the development repo, so I do
not
Post by Brady, Mike
understand how/why the Kolab16 repo would have a later version.
Winterfell does have a later version of erlang, so maybe there is
a
Post by Brady, Mike
version incompatibility?
We have also had to disable Wallace at it constantly (several
times a
Post by Brady, Mike
week on a 50 user system) stopped processing emails.
Post by Roland Kolb
Hi Nik,
due to we have also this performance issue (up to 15min to get
mails) since the upgrade to guam-0.9.2 I think also to owngrade
Guam
Post by Brady, Mike
Post by Roland Kolb
to 0.8.3
Changes in kolab.conf don't help as it helped during the last
performance issue with guam one year ago.
What is the best way for downgrading guam you suggest?
Thanks
Roland
I would think so. Guam will crash and then restart so lots of
spinning wheels on user's screens. It wasn't just the slow
performance for us. We have a mix of systems and clients. Macs
were Ok with Outlook, but Windows PC's (7&10) would not work at
all
Post by Brady, Mike
Post by Roland Kolb
with Outlook. Thunderbird was fine on Windows.
The error messages were misleading (bad CA etc.) and it took two
days to figure out what part of the system was at fault.
Regards
DATE: Saturday, 8 July 2017 at 22:54
SUBJECT: Re: Guam broken in Centos 7
I'm using Kolab 16 on CentOS 7.
Since upgrading to guam-0.9.2-2.1.el7.kolab_16.x86_64, I've had
terrible IMAP performance.
I presume that's the same problem.
Guam is broken in Centos 7 installs so if you've done a yum
update
Post by Brady, Mike
Post by Roland Kolb
and your system is trashed, you need to downgrade Guam to 0.8.3
http://linuxdoku.bplaced.net/?p=401 [1] has the rpm. Remove
Guam
Post by Brady, Mike
Post by Roland Kolb
without deleting dependencies and it will work again
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
------
[1]
https://u2619215.ct.sendgrid.net/wf/click?upn=xyHy6Gye8FSpcIWrVzOzlLWdPGKSw9kKJzMXXa3y9xHy1bRzAQd0AotOyHE2efks_PUFsV5AZSt-2FEXoAPnk9gfI38782PLGBQV9UCRgwrAWVlbDyDP9SARQ0S4-2BV2GQIrvaT9vMpuGFe9Vs6WQmQcX81YvlIpqsqshUErRzhcVy7ZI8FKof1aF7aUo0P5bgYs1nCUpP-2BczbKoAd1PwpYA-2Bb7M18xfBswGM7P2xARpCFPEEcuHAZ-2FJJ8KUS2egmk-2B0muhaBQurcPnChRivllh87Hb8f-2BxuH4uQLQKiTorYAQU-3D
Post by Brady, Mike
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
_______________________________________________
users mailing list
https://lists.kolab.org/mailman/listinfo/users
Matthias Busch
2017-07-12 15:02:18 UTC
Permalink
can confirm, on debian 8 with kolab 16 since the upgrade of guam, the
imap performance in thunderbird is terrible.
it takes a very long time to mark mails as read or see new emails.
thunderbird seems to be taking ages

I can confirm that on two servers and multiple clients

I have no idea how I could even attempt to downgrade guam on debian.

I would like to ask that the problem be fixed with a high priority or at
least, the only guam version will be put back in the repository and the
current (newer) guam will be withdrawn until such a time that the
problem is fixed

Loading...