Dream Host GNU Mailman Nuked
![](https://secure.gravatar.com/avatar/5ba36fe59f305508d6fa09d158f9d457.jpg?s=120&d=mm&r=g)
Dream Host had -- something -- happen over the holiday weekend. As of right now, Tuesday 11/29 at 11:51am EST, I believe that all GNU Mailman lists are still offline.
I'm posting here in case any Dream Host list admins are not aware yet that your lists are likely not working.
The last message my list received was 8:08AM on Sunday (and traffic was unusually slow for two days before that). My list is usually at its busiest on Sundays (semi-retired psychotherapists there as much for socialization as professionally).
Lists are not functioning and admin control panels not reachable. They also had other webserver and email problems Sunday and Monday that seem to be sorting out.
Yesterday they were saying they had a solution in mind. At this moment their status webpage merely says "*[Identified]*Our Technical Operations team has identified an issue impacting Discussion Lists and is currently working on a resolution. We'll update this post with new information as it becomes available."
See: https://www.dreamhoststatus.com
-- Michael
-- *Michael Reeder, LCPC * *Hygeia Counseling Services : Baltimore / Mt. Washington Village location* *410-871-TALK / michael(at)hygeiacounseling.com* *http://www.hygeiacounseling.com - main website. *
![](https://secure.gravatar.com/avatar/5ba36fe59f305508d6fa09d158f9d457.jpg?s=120&d=mm&r=g)
This issue slowly resolved this afternoon and GNU Mailman lists are operational again.
It was Saturday to this afternoon.
I'm pasting in the explanation below -- you only need to read further if you care why they had a service failure. If I understand this right (I might not), they have an archive machine which is a single point of failure that had a power supply blow out. I'm kind of wondering why the load is not distributed with some redundancy and fail-over capability?
Hello,
Please accept our apologies. On Saturday, November 26, one of the
machines within our discussion list cluster had suffered a catastrophic
hardware failure that impacted all discussion lists, both managing and
utilizing. This machine was responsible for the bulk of the data
archived for these lists.
Our technical operations team became aware of the issue when loads
amongst the cluster began to rise and were unable to access the archive
server for list data. After investigation, it was found that the
archival machine was unreachable and required manual intervention to
assess. Once the issue was discovered (bad power supply) they began
discussing the proper paths forward to restore service without any data
loss or data corruption. Due to the holiday season, it was decided to
restore the machine with a repaired power supply and verify data
integrity while planning a future maintenance to fully rebuild and
reimage the machine sometime in January.
At time of writing all machines within the discussion list cluster are
fully operational and list services are working as expected. If you are
still experiencing any issues with discussion lists, please contact
Technical Support for further assistance.
Thank you,
DreamHost Support
*Michael Reeder, LCPC * *Hygeia Counseling Services : Baltimore / Mt. Washington Village location* *410-871-TALK / michael(at)hygeiacounseling.com*
On 11/29/2022 11:57 AM, Michael Reeder -- Hygeia MS wrote:
![](https://secure.gravatar.com/avatar/5ba36fe59f305508d6fa09d158f9d457.jpg?s=120&d=mm&r=g)
This issue slowly resolved this afternoon and GNU Mailman lists are operational again.
It was Saturday to this afternoon.
I'm pasting in the explanation below -- you only need to read further if you care why they had a service failure. If I understand this right (I might not), they have an archive machine which is a single point of failure that had a power supply blow out. I'm kind of wondering why the load is not distributed with some redundancy and fail-over capability?
Hello,
Please accept our apologies. On Saturday, November 26, one of the
machines within our discussion list cluster had suffered a catastrophic
hardware failure that impacted all discussion lists, both managing and
utilizing. This machine was responsible for the bulk of the data
archived for these lists.
Our technical operations team became aware of the issue when loads
amongst the cluster began to rise and were unable to access the archive
server for list data. After investigation, it was found that the
archival machine was unreachable and required manual intervention to
assess. Once the issue was discovered (bad power supply) they began
discussing the proper paths forward to restore service without any data
loss or data corruption. Due to the holiday season, it was decided to
restore the machine with a repaired power supply and verify data
integrity while planning a future maintenance to fully rebuild and
reimage the machine sometime in January.
At time of writing all machines within the discussion list cluster are
fully operational and list services are working as expected. If you are
still experiencing any issues with discussion lists, please contact
Technical Support for further assistance.
Thank you,
DreamHost Support
*Michael Reeder, LCPC * *Hygeia Counseling Services : Baltimore / Mt. Washington Village location* *410-871-TALK / michael(at)hygeiacounseling.com*
On 11/29/2022 11:57 AM, Michael Reeder -- Hygeia MS wrote:
participants (1)
-
Michael Reeder -- Hygeia MS