Gandi IWI Blog

  • Version français
  • English version
  • English version

To content | To menu | To search

Tag - hosting

Entries feed

Sunday 20 December 2009

Datacenter outage- resolved

By Nicolas on Sunday 20 December 2009, 19:43 - alert

We experiment a major outage in one of our datacenter rooms. We'll keep you updated as soon as we have additional information.

Please accept our apologies for the inconvenience

Gandi team

20:26 CET it was a power outage, services are coming back now

20:55 CET Gandi's website is back online, as well as email. The GandiBlogs will return shortly as well as the servers that were affected. We are currently verifying the disks' Raid state

00:10 CET All services are now stable, however we have still received no communication from our hosting partner as to the cause of the outage. As such we are remaining on standby in case of any further incident

09:20 CET 21/12/09 Equinix have scheduled a maintenance window at 10:30 CET to switch over to generator power. There should be no impact to services, but our teams are on alert in case this activity does not go according to plan.

--11:10 CET 21/12/09__ Equinix have not yet effected the switchover. Our teams are still on standby alert while we wait for an update.

11:30 CET 21/12/09 Equinix have now effected this switchover.

Friday 13 November 2009

Network Maintenance overnight 19-20 November

By Leland Vandervort on Friday 13 November 2009, 16:12 - Maintenance

We will be performing a major network maintenance operation during the night of 19 to 20 November in order to further improve our infrastructure.

This maintenance will commence on 20 November at 01:00 AM CET and finish at 07:00 AM CET.

During this time, three connectivity outages are expected, each lasting approximately one hour and will impact most Gandi services.

All hosting operations will be suspended throughout the maintenance period.

We apologise in advance for any inconvenience caused.

Thursday 22 October 2009

Network Maintenance 28-30 October 2009 (Completed)

By Leland Vandervort on Thursday 22 October 2009, 16:30 - Hosting - Maintenance

We will be performing major upgrades to our network distribution infrastructure during the nights of 28 to 30 October. (This is the maintenance rescheduled from 21-23 October) During this time Gandi services will experience a number of intermittent disruptions in connectivity beginning at 1am (CEST) with maintenance activities finishing by 7am (CEST) each day. We will, nevertheless, endeavour to minimise impact to services where possible, but given the scale of the upgrades being performed, most services will encounter at least some disruption. Gandi Hosting will be the most affected during this operation, with an expected cumulative total of approximately 90 minutes of network outage. Other Gandi services will be partially affected.

We apologise in advance for any inconvenience this work may cause.

UPDATE 2009-10-30 05:20 CET: This work has now been completed and brings additional capacity, resilience and performance to the distribution layers of the network. We will be scheduling a further maintenance in the coming weeks to further enhance the features of the new hardware that has been installed. We will of course keep customers informed nearer the time.

Wednesday 21 October 2009

Network Maintenance 21-23 October 2009 (cancelled for tonight)

By Leland Vandervort on Wednesday 21 October 2009, 09:03 - Maintenance

We will be performing major upgrades to our network distribution infrastructure during the nights of 21 to 23 October.  During this time Gandi services will experience a number of intermittent disruptions in connectivity beginning at 1am (CEST) with maintenance activities finishing by 7am (CEST) each day.  We will, nevertheless, endeavour to minimise impact to services where possible, but given the scale of the upgrades being performed, most services will encounter at least some disruption.  Gandi Hosting will be the most affected during this operation, with an expected cumulative total of approximately 90 minutes of network outage.  Other Gandi services will be partially affected.

We apologise in advance for any inconvenience this work may cause.

EDIT : The network upgrades are deferred until a further date.

Wednesday 19 August 2009

Incident on Air Conditioning in Datacentres - Resolved

By Leland Vandervort on Wednesday 19 August 2009, 23:29 - Maintenance

As a result of a problem with the air conditioning in two of our datacentre locations, we have had to halt all services in that location.  This impacts pretty much a quarter of our hosting and all sitemaker services as well as one of our DNS clusters.  We are working with our colocation partners to attempt to restart the services as quickly as possible.

We will keep you informed of developments on this issue.

We apologise for any inconvenience.

23:40:  Our colocation partner is currently migrating to a backup cooling system.

00:40:  The temperature is slowly dropping, and we will being by reactivating part of the services.  Our colocation partner is still working on repairing the cooling system, however, but we do not yet have an estimated time to repair.

07:00: We are now staring to bring services back up in the second datacentre affected by the cooling system failure.  Some services still degraded, however.

07:45: Most services are now available.

08:15: Incident has been resolved.  Any customers still experiencing problems with their servers, please contact support via the online form.

Thursday 23 July 2009

Hosting Storage scheduled maintenance

By flyou on Thursday 23 July 2009, 18:14 - Maintenance

The maintenance will begin :

Start : 02:00 AM CEST Duration : 30 min

Only customers who have received an email are concerned.

We remind you that during this maintenance, which will last 30 minutes, your server will not be stopped. We will nonetheless suspend read/write operations on the disk.

If you use your disks intensively, we recommend that you stop your server, or remove the concerned disks before the maintenance begins.

Please accept our apologies for any inconvenience this may cause.

edit 02:25 Operation has been completed successfully. Affected disks should be functional now.

Thursday 9 July 2009

Partial Hosting Network Disruption

By Leland Vandervort on Thursday 9 July 2009, 17:54 - Maintenance

As a result of a temporary issue on a piece of network equipment, a small portion of our hosting customers may have experienced a brief disruption to the public network connectivity for their servers. This was purely a routing issue for a few minutes. There will have been no impact to the storage network as this issue only affected the public IP LAN for some of the servers.

We apologise for any inconvenience this may have caused.

Thursday 2 July 2009

Electrical problem in one of our data centers

By Ryan on Thursday 2 July 2009, 11:39 - alert

A part of our hosting platform has been impacted by a total power outage in one of our data centers. We are currently working on the problem so that the situation can return to normal ASAP.

We will, of course, provide you with additional information on this incident once we have reestablished the service.

12:20 UPDATE: the power outage is over in datacenter 1. The servers are all restarting automatically, though it will take about half an hour for all of the affected servers to be rebooted, given the large number involved.

13:20 UPDATE: it will finaly take more time for some customer as the hosting backend is highly stressed. We will check that all servers will be back up at the end of the process.

15:30 UPDATE : the servers have now all been rebooted, however, some remain that need an fsck check. We are taking care of that.

Saturday 6 June 2009

Incident on the hosting.

By aegiap on Saturday 6 June 2009, 11:15 - Maintenance

Due to a network equipment failure, some virtual machines are now unreachable. Our staff is currently working on fixing this issue. We apologise for any inconvenience.

12:15 AM : All maintenance are completed.

Tuesday 21 April 2009

Slower creation time for GandiAI servers

By aegiap on Tuesday 21 April 2009, 12:06 - Maintenance

The creation and update of servers using GandiAI is currently taking longer than usual due to an issue with the package server. We are working to resolve the issue.

The technical team.

edit: Following ubuntu gutsy 7.10 end of life and removal from the official repositories, we are setting up a new dedicated mirror. It is currently synching and will be fully available in a few hours.

GandiAI installations will occur at this time with no intervention.

Experts users will benefit from this repository by upgrading /etc/apt/sources.list with the following repositories:

deb http://mirrors.gandi.net/ubuntu-old gutsy main universe multiverse
deb http://mirrors.gandi.net/ubuntu-old gutsy-security main universe multiverse
deb http://mirrors.gandi.net/ubuntu-old gutsy-updates main
deb http://mirrors.gandi.net/gandi/ubuntu gutsy main

edit: problem is now fixed

Wednesday 15 April 2009

Slight Delays in Hosting Operations

By Leland Vandervort on Wednesday 15 April 2009, 09:39 - Maintenance

We will be performing scheduled maintenance this morning on our hosting API.

As a result, during this time, basic administrative operations (creation, stop/start of VMs, attachment/detachment of disks, addition/removal of memory/cpu) may be delayed by between 15 and 30 minutes.

We apologise for any inconvenience caused.

edit: maintenance is now finished and did not delay your operations

Wednesday 18 March 2009

Temporary Disruption of Hosting Services

By Leland Vandervort on Wednesday 18 March 2009, 11:13 - Maintenance

As a result of an anomaly encountered on our hosting management system, purchases and configurations for hosting shares are temporarily unavailable. We are currently working to resolve the issue.

We apologies for any inconvenience caused.

UPDATE: The problem has been resolved.

Friday 27 February 2009

Filer inaccessibility

By Ryan on Friday 27 February 2009, 17:37 - Maintenance

Following a problem on a storage unit, the hard disks of 215 virtual servers have been rendered inaccessible. We are working on a resolution to the problem. The disks will soon be accessible.

Please accept our apologies for any inconvenience this may have caused.

The Gandi team

Tuesday 20 January 2009

Mail: the bug is still there

By Sam on Tuesday 20 January 2009, 12:44 - Maintenance

About a third of our customers are currently experiencing delays in with their mail service.

The reason is unfortunately known: the bug we have already identified is still there, but has taken a slightly different form, and causes the locking of some mail boxes (precisely 160 have been clearly identified).

There are three different situations:

- If you are experiencing a failure of your mail box, there is nothing to do but wait for us to fix your account. This will be done automatically and as soon as possible.

- If you are experiencing delays in your messages reception, please wait around half an hour before trying to get your messages again, reception should not be a problem by then.

- If everything is working correctly for you, congratulations, you are among the majority of our customers for which our fixes were successful.

In all cases, we apologize for the inconvenience. We will keep you informed of the measures taken and of our progress in the task of providing you with the perfect email service (= i.e. normal).