5 ms·
OVH Cloud Storage is inaccessible for 2+ hours
In other news the "VIP support" is enjoying a public holiday.
- openmaze 9y agoHere is the ticket http://travaux.ovh.net/?do=details&id=26480&edit=yep http://travaux.ovh.net/?do=details&id=26480&edit=yep
- clon 9y agoAlso these: http://travaux.ovh.net/?do=details&id=26479 http://travaux.ovh.net/?do=details&id=26479 http://travaux.ovh.net/?do=details&id=26481 http://travaux.ovh.net/?do=details&id=26481 http://travaux.ovh.net/?do=details&id=26475 http://travaux.ovh.net/?do=details&id=26475 The last one seems to have the most up-to-date information. Apparently the Keystone issue is unrelated to the WAW D/C prodding.
- baqs 9y agoswift services are now up
- clon 9y agohttps://www.ovh.ie/public-cloud/ https://www.ovh.ie/public-cloud/ Claims no less than 99.999% SLA: Daily: 0.9s Weekly: 6.0s Monthly: 26.3s Yearly: 5m 15.6s
- mcintyre1994 9y ago> OVH Public Cloud guarantees 99.999% SLA for all available resources. (compute, storage, network) Where the support team detects a lost connection that isn't caused by scheduled maintenance, you will receive damages of 0.5% of the monthly rate per minute of unavailability (up to a maximum of 100% of your consumption for the current month). Clever, they get to schedule maintenance outside of the SLA guarantee so it's entirely meaningless.
- dx034 9y agoBut they usually admit to errors. The public logs in this case confirm that it wasn't maintenance. For OVH (as for any provider) it's better to promise 99.999% and pay if you have an issue than to promise 99.99% and lose clients to AWS.
- toomuchtodo 9y agoMost SLAs don't cover scheduled maintenance.
- onestone 9y agoGood luck claiming SLA refunds from them. Their "support" will stall for months and in the end forget/expire your requests.
- openmaze 9y agoStill users can't admin their OpenStack machines, containers, networks...
- AznHisoka 9y agowhy would someone use OVH for cloud storage when they can use S3? OVH is only good for dedicated servers.
- no_wizard 9y agoAmazon has outages too. Remember when AWS was down for mor than half a day not that long ago? https://aws.amazon.com/message/41926/ https://aws.amazon.com/message/41926/
- codewithcheese 9y agoBecause it is much much cheaper.
- clon 9y agoOVH PCS $0.0112/month/GB AWS S3 $0.0245/month/GB Less than half the price, if you are willing to believe they are comparable products. Actually pretty great performance they offer as well for taking encrypted backups etc. But I think your comment is still generally fair in terms of the quality of the rest of their offerings. I would also add one more strong point for OVH - kick-ass DDoS protection. S3 has had it's own (catastrophic) issues as well. The failure mode for PCS so far has been more irritating, however - it will fail you every 1000'th request or so.
- candiodari 9y agoAlso data transfer pricing is somewhere between 10% and 20% of the Amazon price. That's not 10% off. That's 90% off. https://www.ovh.com/us/public-cloud/storage/object-storage/ https://www.ovh.com/us/public-cloud/storage/object-storage/ https://aws.amazon.com/s3/pricing/ https://aws.amazon.com/s3/pricing/
- vincentkriek 9y agoI am not that into the whole data storage world but for backupping my NAS I looked into Backblaze which is a lot cheaper than both at $0.005/month/GB.
- pfg 9y agoBased on my logs, the outage seems to have started around 9 am UTC, with services recovering around 10:50 am UTC.
- Antwan 9y agoTypical OVH.