New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
Comments
Im also a customer at idrive and i have also a bucket in frankfurt but it was up the whole time.
Message was "Frankfurt region under maintenance" but not all customers were impacted. I guess it depends on the clusters/servers.
Wow. Thanks for the suggestion. I will take a look. Right now i'm using 3 combinations. GDrive, Koofr and Cloudflare R2. It's all free
Mainly use Rclone for backup the websites.
Glad these reliability stats I posted a while ago are being useful!
From what I understand, IDrive outsources their engineering and support to developing countries.
My guess is they're maintaining individual servers manually, and your buckets are only stored on one server in a region. So if a technician screws up a server, it can be days before they get enough complaints to bother fixing it.
More like 30 guys being paid the combined salary of what 3 engineers would be paid in the US.
Given the length of my outage, and other stats I have seen, then yes, they probably have data stored on one server, zero resilience.
Perfectly fine for hobbyist data that is not critical and can wait for a few days to be accessed again, but very far from what they advertise (99.9% availability, which is not possible without resilience, and suitability for live, critical data).
It also makes sense that some servers have had expired certificates on them. It's probably a manual maintenance process for each server.
Confirmed that IDrive E2 is using MinIO on their backend. Found this on LinkedIn.
I once got a server down for less than 8 hours and by the end of it i had suicidal thoughts
Oh, myw's backup stopped since 27 Jul.
I used there Frankfurt site to store 150GB of data and got billed 5 USD some days after - seems like mounting the bucket via s3fs and running borg on it downloaded the whole bucket every day. Free egress only fits if you download less then 3 times of the stored data per month.
not good i was looking at idrive e2 as a contendor for wasabi s3
Someone is reporting that his storage in Virginia has been down for more than a week now...
I wouldn't leave my credit card details with those guys. Best option is to pay by Paypal and then cancel the recurring payment on Paypal, so that they can't take further money after the initial payment is made (assuming buying for 12 months).
Actually there are some alternatives, like Cloudflare R2, Contabo Object Storage and Tebi. I used both Cloudflare R2 and Tebi, it's pretty solid so far.
Are you still having problems? I have been using them for a year in Frankfurt and it's been awesome. Just a couple of minor hiccups so far resolved very quickly.
Ticket please so I can check this. 👌
Do you work for iDrive???
No, he's the owner of myw. I assume myw use idrive for its backup (CMIIW)
No hurry, let's wait until iDrive solves it.
Oh, I see terminal and x-ray in direct admin panel, so cool!
I just visited the site once about s year ago and understood right away that it was trash
Last few years I’ve noticed with all host fuck ups on here it’s like I’ve trained some internal AI to know when something is bad
Can’t say why
And for those of you who use this for backups, come on, use Backblaze/AWS or something
If you use it for very regular downloads, Cloudflare R2 is very appealing
Obviously not as cheap, but at least you can sleep well at night (and I’d still revommend not only having the backup at one place)
True, having one backup in one place is wrong. But only one backup might be good in some cases, as long as this backup is extremely reliable and you get notifications when something goes wrong, so that you may create another backup solution for your data afterwards (outside existing provider).
The problem in this case is not the backup or the concept of backup. The elephant in the room is iDrive who disappointed its customers and did not even notify them with regards outage or about reliability beforehand (like SLA warning when ordering).
As @emgh pointed out, maybe now is a good time to think about other providers for backups. This is low-end market where offers are welcomed. It seems offers for storage are highly needed now, due to iDrive mistakes.
Correct, multiple backups are always useful. Also, we must keep our backups in different locations.
Yep, we provide all that!
Correct, very affordable to us. We'll also submit a ticket with them to try to rush things.
Interesting, keep following this thread. What I wonder is if the crazy prices and promotions on the iDrive could be the reason for their increased downtime.
As well they don't seem to be stating how redundant their data is - according to someone's digging (I can't remember which one) they are based on minio-processed S3 clusters, so redundancy should be an important metric, not just claiming a string of 9's for data reliability, right?
It is best to be a good technician and assume the worst. I consider at least 2 backups. The moment one backup fails (whether is cold storage or some cloud storage) make another backup. In this case if iDrive has failed after multiple days, make another backup with some other cloud provider; just to be safe until iDrive gets their business in order.
I cry everytime when my credit card is charged by Backblaze but I get a sound sleep at night, knowing my data is safe (mostly family photos).
Backblaze may seem to be costly but it's worth every penny for the amount of service it's delivering.
I never had any problem with them. Last I contacted them was due to fault in their android app.
Their poor uptime (across many regions) is just by design: they probably store customer data on single servers (the very unique end-points provided is an indication), and servers go down. They have no resilience, so when a server goes down, it takes days to rebuild.
If it was advertised as the product it actually is, then wouldn't be an issue. Resilience does cost money, and they are cheap.
The issue is that they deceive people advertising their service as an enterprise-grade solution with "99.9% availability" that can be used for storing "critical and live data". Check their website.
They perfectly know it's not true, by design.
Vultr NJ object storage has been down for most of August too, and they definately arent using a single server.