I’m curious if there is a status page hosted on separate infrastructure for times when various Purism sites go offline, such as what happened about 16 hours ago. I think a fediverse account should also be used to address these situations by alerting the community and directing people to the status page.
Also, I’m curious why it happened this time. Does anyone know for certain what the cause of the downtime was? I can only guess, but I don’t want to speculate in this post.
I don’t think so. I think Matrix is the way to get info from Purism if the web site is down.
If any web site is having a planned outage then it should be announced in advance on the web site itself.
No, but you may be interested in reading a very similar thread:
You did just remind me of Uptime Kuma, which is a tool I have seen used by a few privacy front-end administrators:
To the best of my knowledge there is not one for the forum, nor for awe-sim, and the one for librem.one https://status.librem.one/ hasn’t been updated since 2020 despite multiple outages. There’s a history of not puting forth the effort here and I don’t see that changing anytime soon unfortunately.
I forgot to mention a very important point.
You should not trust any Librem Social accounts with this critical vulnerability still not being formally addressed by Purism.
Yeah, I remember that. I’m guessing the same thing happened this time but on a smaller scale, since there has been no word from Purism about site maintenance. Then there’s the question of motivation, if this is indeed what happened. I hope Purism will comment on it.
A financial oversight where a relatively inexpensive hosting invoice was missed, causing an outage, and was resolved once the appropriate staff was able to respond. Purism is a fully-remote company and is lacking some staff/operational continuity policies, which I’m working to improve as bandwidth allows.
Server upgrades still need to happen and is a documented systems issue. This vulnerability has been patched, however.
Thank you for the information.
The main site “Puri.sm” was still up and running. Maybe the news page is a good place to inform visitors of the outage.
About those policies to fix: just noted too, as was also noted in the case of the previous major downtime, why there again was no notification on official website or official company Mastodon (Purism (@purism@librem.one) - Librem Social) or even any notes with #purism tag about it by other users.
Next week I go for a DR planning exercise at my job in Nebraska. The kind of offsite DR plan in case a tornado hits the computer room, other manufacturing sites in other states can continue to run and use the business system.
There is a instatus.com uptime/outage reporting site. It includes incident, maintenance, and custom incident/maintenance type templates for notices. Announcement and contact information could be in public or private view, depending on the security requirement of the website. I am not sure if its purpose is strictly technically, but it’s login method includes a OTP magic link, which is a preferred way compared to just password. I can’t say if there’s MFA, but that may change in the future.
I am not sure how secure is this service, but my confidence level is pretty good given the fact that it is just a website status reporting service.
You can self-host your own monitoring service instead. I use Uptime Kuma for my OVHcloud VPS to monitor various privacy front-ends (among other services) at the moment:
I already mentioned this software earlier in the topic, but I ended up following through with its deployment.