[Adminsysters] TECH INFRA - PREP UpGRADE - issue [mur.at #12287] backup

Ada LaNerd ada at hackstub.netlib.re
Thu Oct 17 15:34:23 CEST 2024 

Hello,

You can find the extensive summary of our last operations in the pad if 
you're interested, we tried and document it as much as we could so 
anyone of you can learn some things out of it.

By the way, mara noted that tinc, the mesh VPN (decentralized virtual 
private network) protocol we intended to setup, was discontinued and 
also learned that a new VPN was recommended nowadays but couldn't 
remember its name. After digging a bit, I found out about Netmaker 
<https://www.netmaker.io/>, which is based upon Wireguard and looks 
good. Is it the right one?

Would anyone have some recent experience in such VPN solutions?

bye,

Ada

Le 15/10/2024 à 14:31, Ada LaNerd a écrit :
>
> Hi everyone!
>
> Apparently tomorrow 14:00 CET is the most suitable slot for this 
> session. The previous link for the BBB was 
> https://visio.futuretic.fr/360 and the pad is there : 
> https://eth.leverburns.blue/p/borg
>
> I'm not sure the status of the backup is very clear: is it done or not 
> yet?
>
> The goal of the session would be to end the upgrade to Debian bookworm 
> before we can install tinc right?
>
> hugs for those who want some!
>
> Le 10/10/2024 à 13:35, bolwerK a écrit :
>>
>> hi
>> @ igni did you already start the backup solo
>> thought we were agreeing on a date :: 
>> https://transitional.anarchaserver.org/date/studs.php?poll=9hZvhKvLFK7Lu3Ug 
>>
>> and it seems  wednesday 16/10 we can be more people present
>> so can we do it all together that moment  - as in the meantime  we 
>> received this message from ralph of [mur][at]
>>
>> xm
>>
>>
>> -------- Forwarded Message --------
>> Subject: 	Re: TECH INFRA - PREP UpGRADE - issue [mur.at #12287] backup
>> Date: 	Wed, 9 Oct 2024 14:46:43 +0200
>> From: 	Ralph Wozelka via RT <amt at mur.at>
>> Reply-To: 	amt at mur.at
>> To: 	info at ooooo.be, mara at multiplace.org
>>
>>
>>
>> it looks like the transfers broke off pretty early in the process!
>>
>>> murmin at hamster/backup/images/adelejean »ll
>>> total 1,1G
>>> -rw-rw-r-- 1 murmin murmin    0 Okt  3 17:53 adele01.img.lz4
>>> -rw-rw-r-- 1 murmin murmin  350 Okt  4 23:40 jean01.img.lz4
>>> -rw-rw-r-- 1 murmin murmin 1,1G Okt  3 15:32 test.img.lz4
>>> murmin at hamster/backup/images/adelejean »
>>
>> I will write a script which will loop back and restart the receiving 
>> server-side process!
>> It will *keep* all files from previous attempts so that nothing can 
>> go amiss :-)
>> it will name the backup files like "adele01.img.lz4.202410091415", ...
>> and "jean01.img.lz4.202410091415", ...
>> each new attempt will create a separate file! :-)
>>
>> use the lines as before again to commence transfer:
>> (I added -N for netcat so it will terminate after the end of file is 
>> reached. I noticed it just keeps going wailting for more input otherwise)
>>
>> for adele, port 12346: dd if=/dev/<your-dev-on-adele> | lz4 | pv | nc 
>> -N  hamster.mur.at 12346
>> for jean, port 12347: dd if=/dev/<your-dev-on-jean> | lz4 | pv | nc 
>> -N hamster.mur.at 12347
>>
>>
>> let me know how it goes!
>>
>> so long,
>> r
>>
>> If the connection breaks it will restart the listening servers on the 
>> same tcp ports ready for transfar
>> On 03.10.24 17:59, Ralph Wozelka | NOC - mur.at wrote:
>>>
>>> I'll prepare for 2 separate transfers now!
>>> install lz4 and pv on your servers.
>>>
>>> try some transfer locally to make sure the tools work.
>>>
>>> then you should run this line:
>>>
>>> dd if=/dev/<your-dev-on-adele> | lz4 | pv | nc hamster.mur.at 12346
>>> dd if=/dev/<your-dev-on-jean> | lz4 | pv | nc hamster.mur.at 12347
>>>
>>> and I start 2 corresponding server on hamster *now*, which will 
>>> store the streams to local files there:
>>>
>>> /backup/images/adelejean »nc -l -p 12346 | cat > adele01.img.lz4
>>> /backup/images/adelejean »nc -l -p 12347 | cat > jean01.img.lz4
>>>
>>>
>>> --> the server side is up and running *from now on* and you can 
>>> start your transfer whenever you're ready.
>>>
>>> *note:*  the server processes will exit after your client 
>>> terminates. so if some error occurs on your side, I will need to 
>>> restart the server-side processes.
>>>
>>> it is the fastest way of transferring the data via the local network 
>>> (it is not encrypted - so only good for our local network. it would 
>>> be a rather poor choice for across the public internet)
>>>
>>> so long,
>>> r
>>>
>>> On 01.10.24 23:43, info at ooooo.be via RT wrote:
>>>> Tue Oct 01 23:43:50 2024: Request 12287 was acted upon.
>>>> Transaction: Correspondence added byinfo at ooooo.be
>>>> Queue: noc
>>>> Subject: TECH INFRA - PREP UpGRADE - issue [mur.at #12287] backup
>>>> Owner: rwoz
>>>> Requestors:info at ooooo.be,mara at multiplace.org
>>>> Status: open
>>>> Ticket <URL:https://rt.mur.at/Ticket/Display.html?id=12287 >
>>>>
>>>>
>>>> Hi  great for the enthousiasm.
>>>> Is raw  disk spaec around 500 gig
>>>>
>>>> $ df -h
>>>> *adele */dev/md0p2      876G  167G  665G  21% /
>>>> *jean* /dev/mapper/jean-data 875G  357G  474G  43% /
>>>>
>>>> and do you have a little more explanation what you expect from us ?
>>>> "/netcat socket on hamster"
>>>>
>>>>
>>>> xm
>>>>
>>>> sounds good
>>>> On 30/09/2024 17:10, Ralph Wozelka via RT wrote:
>>>>> hey systers!
>>>>>
>>>>> how big are the disk(s)?
>>>>> we could do it with netcat!
>>>>> I'd open a netcat listening socket on hamster, our backup server, and
>>>>> you could pipe the whole block device there via netcat...
>>>>>
>>>>> so long,
>>>>> r
>>>>>
>>>>> On 16.09.24 18:20,info at ooooo.be via RT wrote:
>>>>>> Mon Sep 16 18:20:09 2024: Request 12287 was acted upon.
>>>>>> Transaction: Ticket createdbyinfo at ooooo.be
>>>>>> Queue: noc
>>>>>> Subject: backup
>>>>>> Owner: Nobody
>>>>>> Requestors:info at ooooo.be
>>>>>> Status: new
>>>>>> Ticket<URL:https://rt.mur.at/Ticket/Display.html?id=12287 >
>>>>>>
>>>>>>
>>>>>> hello mur.
>>>>>> we write you / the technical support -
>>>>>> cc Andreas
>>>>>> to see if you can follow up our request?
>>>>>>
>>>>>> on 12/09 we wrote ANdreas --
>>>>>>
>>>>>> -------------------
>>>>>>
>>>>>> Dear Andreas
>>>>>>
>>>>>> hope you are doing well.
>>>>>> We want to upgrade adele server, but our backup server has some 
>>>>>> issues
>>>>>> so we are wondering if mur could make a full disk backup for us?
>>>>>>
>>>>>> -------------------
>>>>>>
>>>>>> thanks for taking it in consideration
>>>>>> xm
>>>>>>
>>>>>>
>>>>>>
>>>>>> thanks xm
>>>>>>
>>>>>>
>>>>>>
>>>>>>
>>> -- / NOC - mur.at
>>> / Network Operation Center / Systemadministration
>>> / +43 (0)681 816 45 308
>>> / Di-Fr 12:00-14:00 (Support)
>>
>> -- 
>>      / NOC - mur.at
>>     / Network Operation Center / Systemadministration
>>    / +43 (0)681 816 45 308
>> / Di-Fr 12:00-14:00 (Support)
>>
>>
>> _______________________________________________
>> Adminsysters mailing list
>> Adminsysters at lists.genderchangers.org
>> https://lists.genderchangers.org/mailman/listinfo/adminsysters
>
> _______________________________________________
> Adminsysters mailing list
> Adminsysters at lists.genderchangers.org
> https://lists.genderchangers.org/mailman/listinfo/adminsysters
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.genderchangers.org/pipermail/adminsysters/attachments/20241017/3a906f8f/attachment.htm>

More information about the Adminsysters mailing list