cover photo


Is now down.
Mike Macgirvin
Father of all, we pray to you for  and for all those whom we love but see no longer. Grant to them eternal rest. Let light perpetual shine upon them. May their soul and the souls of all the departed, through the mercy of God, rest in peace. Amen.
attention all ships

After four years is shutting down. This will happen on Monday 09 April 2018, sometime in the morning (CEST (UTC+2)).

Anyone with content still on there that they would like to get off, this is your chance.

My identity is now on my Hubzilla instance: elmussol. It is possible that I have missed moving some of my Diaspora and Friendica connections. If I have missed you and you'd still like the connection, please connect there.

els Mussols is also on the new server, so you don't have to lose the egg count.
James Lamentus
  last edited: Tue, 03 Apr 2018 23:21:18 +0200  
after 4 years and  a closed site/hub   elmussol (channel) is still  here... And you ( that use other software/network) ?
You #GotZoT ?
No ?! Ahi ahi ahi :facepalm
Michael MD
still thinking about adding some zot features to the main spraci site here
(obviously the hubzilla instance already has zot but thinking it would be nice to add openwebauth and federation features directly into
- still interested in getting federation for events listings going)
Redmatrix Photo Migrator 1.0

@Hubzilla Support Forum+ I have this plugin enabled as I would like to get as much content as possible off The documentation for this plugin is spartan:

Migrate photo albums from Redmatrix to Hubzilla

Author: Mike Macgirvin

Maintainer: Mike Macgirvin

If I get a clue on how to use it, I might very well document it.
Mike Macgirvin
Instructions: Visit /redphotos on your site while logged in. Fill out the form and submit.
That worked. Thank you.
redfiles worked in exactly the same way obviously.
so what's the deal with all the old posts?

 Xerta last edited: Wed, 21 Mar 2018 00:07:51 +0100  
Why, oh why, oh why are all my old posts appearing in your timeline? As one of the last people on earth to running a RedMatrix instance I have been slow at Hubzilla adoption. Not for any other reason than I already work to hard, sysadmin makes my head hurt and RedMatrix still works brilliantly.

But Parlementum (where my clone is) is closing down and I have received enough kicks in the arse now to make the Hubzilla happen. Unfortunately I suspect there are too many database changes between latest RedMatrix and Hubzilla to make a cloning to my new instance successful. So I am making sure that all my stuff is on Parlementum and using that export to clone to my new instance.

So, tough ;-)
the spamming is fine :) it's more  about doing away with headaches innit ... some times i have felt secretly relieved when loosing some kind of massive archives of information ...
Christian Vogeley
Hoping for the rebirth of the #eggcount
Michael MD
regarding database changes ,,, had the same problem . but set this hubziilla instance up a while ago ...

though up until only a couple of weeks ago the old red instance here was also still working...

before the big server crash (not red or hubzilla related .. only clues I found were a few lines in syslog which looked like a hardware problem - followed by mysql errors that looked like it was having trouble reading the hard drive ,,,

if it was hard drive failure not the first time that happened to me .. but given that was 15 yeas sgo not looking like "often" or wierd more like just "shit happens". ..

that vps wouldn't restart so now on a new debian 9 vps slowly putting things back together ..

getting the hubzilla instance back up was nice and easy btw..
just copied across the files across from the old debian 6 drive and and it pretty much just worked
including the database (nice feature of myisam :-))  ..  probably not the recommended way but quck and more up-to-date data than I could get from a few-days-old
backup files that would have taken ages to upload .. and it being mysql on the old vps and mariadb on the new one didnt matter ,
no signs of trouble there..

so pretty much copied it all across and it just worked ..

wishing more other things could more often be that easy :-)

I havent restored the old red instance though probably not harder to do probably not needed as I've been using the hubzilla instance more for a while now anyway.
(probably due for an update soon though)

non-display of post

 Xerta last edited: Wed, 21 Mar 2018 01:06:56 +0100  
@Hubzilla Support Forum+ There are a series of posts that don't display for me (and also don't get removed from my unread posts list when selected). This is on RedMatrix so there may be nothing to do. An example: git post from Mike.

I suspect it's the,2008:Grit::Commit bit of the url that it's choking on.
My issue is different from the urlencoding, but it is Redmatrix only, so I do not care so much about it.
I have problems with posts from the following channel:
Before I had a link in the notification like and just resulted in an empty page. Now it is urlencoded and produces a 404. When I click on "Link to source" it goes to
On a hubzilla host everything looks fine and the notification link works (/display/b64.aHR0cHM6Ly9oYW1idXJnLmZyZWlmdW5rLm5ldC8_cD0zMzQ0)
The git pull has been done, but I don't have a new post to check yet. I will report when I have.
Mike Macgirvin
and produces a 404.

I was worried about that. Unfortunately converting foreign message-ids to base64 is a fair chunk of work and outside the scope of my current involvement with redmatrix - which has been essentially in a vegetative state and on morphine drip for the last year.

If you're interested in a quick and dirty fix: you can look at my last checkin there (1 line) and undo the urlencode and do a limited conversion to base64 just for notifications without touching anything else. In mod/display look for that encoding specifically and decode if it has been applied. A more complete solution would be to just base64 message-ids containing punctuation chars. I think you still have git repo rights there. PSA

 Xerta last edited: Wed, 21 Mar 2018 10:52:26 +0100  
My cert runs out in about three hours. It will not get sorted until tomorrow, probably afternoon (European time). Stuff may happen or not in the intervening period.
Mike Macgirvin
That's OK. We know where to find you.
  last edited: Wed, 21 Mar 2018 10:51:53 +0100  
And we are back I believe.

 Xerta last edited: Wed, 21 Mar 2018 12:20:34 +0100  
Yes I can but can't like/dislike or comment.

James D MorganJames D Morgan wrote the following post Fri, 23 Oct 2015 19:42:59 +0200
All right, I switched to a hubzilla doodad... Can anyone see this?
failed update 1148

 Xerta last edited: Wed, 21 Mar 2018 12:47:05 +0100  
@Channel_One+ @The_Red_Matrix_Support_Channel+  When I get one of these I usually know how to fix them, but I am confused by this one:

Failed Updates

From update.php:

function update_r1148() {
    $r1 = q("alter table likes add i_mid char(255) not null default '' ");
    $r2 = q("create index i_mid on likes ( i_mid ) ");
    $r3 = q("select likes.*, item.mid from from likes left join item on likes.iid =");
    if($r3) {
        foreach($r3 as $rr) {
            q("update likes set i_mid = '%s' where id = $d",

$r1 successfully completed, $r2 was where it fell over but I have applied this manually. I don't know how to deal with $r3. I get:

MariaDB [redmatrixdb]> select likes.*, item.mid from from likes left join item on likes.iid =;
ERROR 1064 (42000): You have an error in your SQL syntax; check the manual that corresponds to your MariaDB server version for the right syntax to use near 'from likes left join item on likes.iid =' at line 1

Guidance please.
Mike Macgirvin
Yeah that's the best way out. It isn't a huge problem if it never runs, the purpose is to make profile likes easily exportable to clones. There aren't more than a dozen or two of these in the network. You're welcome to have a go at it. If you do make a new update for it, remove those lines from this update so it doesn't get executed twice, though that probably wouldn't hurt anything either. It's all pretty low impact.
Pull requested.
  last edited: Wed, 21 Mar 2018 12:43:36 +0100  
All that worked. Thanks @zottel and @Mike Macgirvin.
a new perspective

 Xerta last edited: Wed, 21 Mar 2018 13:30:10 +0100  
@Channel_One+ Worth sharing from a new person's perspective.

Michael McKinseyMichael McKinsey wrote the following post Fri, 17 Jul 2015 23:16:51 +0200
Redmatrix vs Hubzilla branding (and my impressions of the project whatever it is called)
On the general discussion of this topic of names (hopefully someone will link this to somewhere it is in context somehow) - my wife who is not exactly in the know about such things as OSS projects, said she liked the sound of redmatrix more than hubzilla.  I personally am probably more of your target market (I am a multi decade sysadmin and dev planning on using redmatrix as the basis for authentication and communication for a project I am working on and I think it is perfect), and the name makes little difference to me.  It is filling the space where I was dreading having to implement OpenAM or something similar and something like Exo, and then try to figure out how to get cross domain authentication working, which I have spent countless hours messing with, and in 1 day r# fills the rolls as well or better than those would have regardless of if they ever worked as designed (which they did not).

One thought I have is there are so many zillas I think it is almost meaningless at this point, where matrix does mean something and represents what at this early stage of my involvement, I think the project is.  I guess there is some history you are trying to escape, but just one on one, redmatrix vs. hubzilla, I think redmatrix is a much better name, and I was not aware of the history.

Whatever you call it, this protocol, and project, is an absolute oasis for me, as for what I am working on, r# out of the box this takes care of several big pieces that were giving me fits trying to implement other ways.
grab your slippers, cardigan and pipe

 Xerta last edited: Wed, 21 Mar 2018 13:45:18 +0100  
@The_Red_Matrix_Support_Channel+ Many years ago when my beard was just Kropotkin-like and not dreaded and I first did my own Friendica instance, I remember that a member fscked up registering and then couldn't register correctly because his name had already been used despite deletion. Is it the same way with Red? If so, what DB poking around do I need to do?
Thomas Willingham
  last edited: Thu, 11 Jun 2015 01:54:03 +0200  
Sorry, I meant Paul - what when wrong for his member, because we can probably fix it.

I understand the problems you had - there are lots of edge cases where we do filter on UID (all your pconfigs for starters).
Mike Macgirvin
I highly recommend using another nickname unless you're a developer and willing to put up with strange issues that may not be possible to resolve. It can cause subtle bugs which are related to having incorrect data at other locations besides your own and are therefore very difficult to track down.

 Xerta last edited: Wed, 21 Mar 2018 13:49:18 +0100  
I am having on of those "I know the setting exists but I can't find it" moments. I have cloned a channel from one hub to another (take a load off friendicared). I can see the button to change the primary at cloning time; but, post-creation, where do I change which hub is the primary?

Thomas Willingham

No idea if or where it's linked in the UI.
That worked, thanks.
potential downtime

 Xerta last edited: Wed, 21 Mar 2018 14:09:17 +0100  
The SSL certificate for runs out tomorrow (02 June). I have been promised some money later today by someone who owes me, but on previous history I am not confident that will happen. I should definitely have money tomorrow, so potentially we may be offline for a day. Sorry.
 from Diaspora
Roger that. Another freebie trick I've used in the past is to get a free 90 day trial cert from the likes of comodo etc. Pain in the arse having to do it every few months but it's a decent cert at a decent price for only an hours work or so ;)
Hey, we're poor and not for the first time in the history of the world someone hasn't paid me on time. They got threatened with their email going offline tomorrow afternoon if the money isn't in my bank tomorrow morning. We'll see.
  last edited: Wed, 21 Mar 2018 14:08:20 +0100  
We are back up. If anyone has problems let me know.
tell me about cool pages

 Xerta last edited: Wed, 21 Mar 2018 18:52:04 +0100  
To do my part in expanding the number of useful and interesting channels, I've created @pages+ -- the idea being that we share our cool and interesting RedMatrix pages. More importantly perhaps share how we did them.

I've set myself the task of converting the content of (presently Drupal 6) into web pages associated with @els Mussols , so I also declare blatant self-interest.
Red on GNUSocial

 Xerta last edited: Wed, 21 Mar 2018 18:53:54 +0100  
Conversations about and touching Red are happening on GNUSocial. @mmn asked the following question in this thread (which I couldn't answer, so am looking for guidance):

Has #zot been peer reviewed by experienced cryptographers and security specialists yet? I had some glances on the protocol spec but didn't feel very comfortable. But I'm not a security or cryptography specialist either, just a paranoid conservatist.
Mike Macgirvin

"Let any one of you who is without sin be the first to throw a stone...”
-- John 8:7
Mike Macgirvin
I just find it hilarious that we get so much critique from statusnet/gnusocial. Our network deletion model isn't perfect. they're right. It's not. They've got none at all. It's also possible to create malicious edits if you allow edited posts. They're right. They don't let you fix your typos at all. And since you can't delete them, you just live with every mistake you've ever made. Lovely.

There is no perfect solution to these problems in a decentralised world, so instead if you can't have perfection, nobody can have anything - even a solution that works just fine more than 99% of the time.

Let's put this into perspective. You can't build a perfect car or airplane that will never crash and never cause an injury. So nobody should ever be allowed to drive or fly? No - you do the best you can and you try to make it better.

Next it's our security and privacy implementation. It is pronounced flawed because somebody who isn't an expert took a quick look and didn't feel comfortable - whatever that means;  and suggests we have it peer reviewed/audited.  Again, let's set the context - this is from somebody who is comfortable using statusnet/gnusocial. Our privacy implementation isn't good enough for them. Does anybody else see the irony? Stop it, you're killing me.
Exactly. And $largeamountofmoney is the point. I always wonder where people who ask this question think this would come from.

 Xerta last edited: Wed, 21 Mar 2018 18:59:55 +0100  
Do the page and read it all.

Mike MacgirvinMike Macgirvin wrote the following post Thu, 02 Apr 2015 01:03:51 +0200
Red Matrix


We're artists and musicians. Intellectuals, geeks, and nerds. Some call us outcasts, but that isn't completely fair. We just look at the world a bit differently than everybody else. Some of us have completely different ideas about gender identity and relationships. Others are too busy improving the world to listen to others complain about it endlessly. Some of us are atheists living in deeply religious families and communities. Others are deeply religious and spend our lives surrounded by atheists. Some of us are just shy.
OpenID success

 Xerta last edited: Wed, 21 Mar 2018 19:18:56 +0100  
Today I successfully commented on a blogspot blog using my Red OpenID credentials for the first time, very impressive. Thanks @Mike Macgirvin (and anyone else who coded this into existence).
 from Diaspora
That's pretty cool :)
  last edited: Wed, 21 Mar 2018 19:20:36 +0100  
Ain't it just.
mildly confused

 Xerta last edited: Wed, 21 Mar 2018 21:23:55 +0100  
If I get an email that says:

The Red Matrix developers released update 1131 recently, but when I tried to install it, something went terribly wrong. This needs to be fixed soon and it requires human intervention. Please contact a Red developer if you can not figure out how to fix it on your own. My database might be invalid.

The error message is 'Update 1131 failed. See error logs.'.

But dbsync page says:

No failed updates.

Which do I believe? Or do I need to do poking around in the db?
Does your webserver have CREATE TABLE and ALTER rights on the database?

ALTER was missing and has now been added, if CREATE is the same as CREATE TABLE, that was already there.
Oh, and thank you :-)
Mike Macgirvin
de nada
oh dear

 Xerta last edited: Wed, 21 Mar 2018 23:22:52 +0100  
@The_Red_Matrix_Support_Channel+ Just did a git pull on red.elsmussols and I have no content in Channel Home and Your Matrix. I last pulled on 28 Dec 2014. Additional info as I have it.
@! and PM

 Xerta last edited: Thu, 22 Mar 2018 00:12:51 +0100  
@Channel_One+  What's the difference?
nginx/ssl setup

 Xerta last edited: Thu, 22 Mar 2018 00:13:49 +0100  
Today's the day and I am just fiddling with the nginx config file for red. My SSL cert is a proper one from gandi and here are the contents of /etc/nginx/ssl:

root@pendle:/etc/nginx/ssl# ls -al
total 24
drwxr-xr-x 2 root root 4096 Oct  6 13:27 .
drwxr-xr-x 6 root root 4096 Oct  6 13:26 ..
-rw-r--r-- 1 root root 1667 Oct  6 13:27 GandiStandardSSLCA.pem
-r-------- 1 root root 1769 Oct  6 13:27
-rw-r--r-- 1 root root 1054 Oct  6 13:27
-r-------- 1 root root 1704 Oct  6 13:27

So this line is easy:

ssl_certificate_key /etc/nginx/ssl/;

But I am unsure how this line should read:

ssl_certificate /etc/nginx/ssl/;
 red  tech