Folksy maintenance mode between 08:00 and half past

Hi — we need to go into maintenance mode for around 15 minutes this morning, we’re thinking somewhere between 8 and half past.

This is to push some improvements and updates that we hope will totally be worth it.🙂

Thanks for bearing with us in the meantime,

Doug.

UPDATE: We’re still running tests on the code to be deployed, so this is behind schedule a little, maybe by 15 minutes.

Certificate issues fixed

Hi — we’re checking through things to see if there are any other issues generated by this, but I thought I’d put a quick update here — the site’s back up.

Thanks for bearing with us, it really is appreciated — I’m off back to checking things with a fine-toothed-comb, now!

Doug.

Folksy’s certificate issues

Hello — we’re really sorry, we’re having issues with our certificate that we’re working to fix as quickly as possible.

We’ll let you know more when we know more.

In the meantime, please bear with us.

Thanks,

Doug.

More updates to the search index

Hi — we’ve been looking into fixing a search-related bug, and have a fix for it that we’ll be pushing out in the next hour or so.

Whilst we’re pushing it out, some people might notice some issues with search. We’re hoping not, but it is a possibility.

In the worst case, we’d have to go into maintenance mode whilst we fixed the search errors, but (without tempting fate!) we don’t think it will come to that.

But I thought I should post this headsup that we’re tinkering with a few things up, to be on the safe side!🙂

Thanks for your patience whilst we’re fixing things,

Doug.

Folksy outage, June 16th 2016

Hi — just a quick precis of the outage we had, today.

The site became unresponsive around 14:00, and had difficulty for around half an hour.

We investigated, and tracked the issue down to one of our load balancers, whereupon we got in touch with AWS support, who host our load balancers.

They were brilliant, as ever, and we were able to identify the issue with them (it was a hardware problem), and fix it, in fairly short order.

Obviously we’ll now be monitoring to make sure everything is behaving as it should be.

We’re sorry for any inconvenience caused, and would like to say thanks again to everyone for bearing with us.

Thanks ever so much,

Doug.

Folksy outage currently being investigated

Hi — just to let you know that we’re investigating the current Folksy outage.

Thanks for bearing with us,

Doug.

Some PayPal issues on Android devices

Hi — there’s a problem with some PayPal transactions on Android devices, currently.
The issue is that buyers are being redirected to a non-existent page when selecting to complete the payment via their PayPal app on Android devices using the latest version of PayPal. If the buyer chooses to use the PayPal website instead of the app, then everything proceeds just fine.
PayPal are aware of this bug, and assure us that they are working on a fix as quickly as possible.
We’re sorry for any inconvenience in the meantime.
Thanks,
Doug.

Image issues yesterday and today

Hi — I just wanted to say a big sorry about the image issues we’ve had yesterday and this morning. People were unable to upload images, and a growing number of existing images were struggling to serve.

There are a suite of checks that run constantly to check for service issues, and I receive automated alerts when any of those fail. However, in this case, I received no alerts. This was because the issue was a new one to us. The image servers were running perfectly, and the cache *appeared* to be running perfectly, so nothing got reported.

It was only upon investigation this morning that I discovered that the cache, whilst working for cached images, wasn’t sending requests for new images to the image servers for processing. This was fixed, and now images should be running fine again.

However, the image service was in this broken state for most of Sunday, and for that I can only offer our most sincere apologies. We’re looking at improving our automated checks so that this issue (and any like it we can think of) would be caught in the future, so that the alert can be sent to me that something needs fixing.

Again, we’re really sorry that this issue lasted as long as it did, and I hope at least that you’re reassured that we’re doing our utmost to ensure it doesn’t happen again.

Thanks,

Doug.

Folksy outage at 17:20

Hi — we had an outage at 17:20 that lasted for around half an hour, whilst we investigated and fixed the issue.

The issue turned out to be a few queries that were really dragging the database down. These queries were called from the homepage, so had a bad effect on the performance of the whole site, rendering it all but inaccessible to some users.

We’ve done a quick fix on one of these queries, and removed the calls to the others, and the site is back up and performant, now. We’re looking into fixing the queries so that this shouldn’t happen again — the investigation of this issue has shown us a pretty good way to approach this, and we’re on with this as our next job.

We’re really sorry to anyone who couldn’t access the site whilst we were having these difficulties, and do apologise for any inconvenience. And offer, as ever, our thanks for your patience whilst we fix these issues that crop up from time to time.

Thanks,

Doug.

Maintenance mode from 06:45 this morning

Hi — we have a deploy that requires rebuilding the shops search index this morning, which will require us to go into maintenance mode for around 10-15 minutes at 6:45am.

We’re sorry for any inconvenience this causes, and will get it done as quickly as possible.

Thanks,

Doug.

Follow

Get every new post delivered to your Inbox.

Join 49,342 other followers