Thanks Willi for making all that clear. We spun up a brand new 16.04 LTS from here https://cloud-images.ubuntu.com/locator/ec2/ and were able to reproduce the error there also. So now its clear that It has nothing to do with the ArangoDB image in particular, but rather aws and perhaps Ubuntu.
Thanks Soren On Wednesday, August 31, 2016 at 5:11:41 PM UTC+2, Wilfried Gösgens wrote: > > Hi, > we basically just install an UBUNTU 14.04 plus add some more packages - > and arangodb in the end. > This is done using packer and ansible. > You can have a look at this over at: > > https://github.com/arangodb-helper/ami-appliance > > Cheers, > > Willi > > On Friday, August 26, 2016 at 11:08:47 AM UTC+2, sdy7 wrote: >> >> Hi all >> >> We are using ArangoDB on the images published in AWS Marketplace. After >> our own few customisations via apt-get plus making a new AMI, instances >> launched from this has very bad write performance on /vol/data. Then after >> a stop+start the problem goes away. A little write test program writes a >> file of 280MB. It takes more than 50 seconds, and only 2 seconds on an >> instance that works like it is supposed to. >> >> Also mentioned here: >> https://forums.aws.amazon.com/thread.jspa?messageID=737841򴈱 >> >> How to reproduce the issue: >> 1) launch a fresh ArangoDB instance from AWS Marketplace (c4.xlarge). >> 2) run apt-get update and upgrade. >> 3) Make a new image and launch a new instance. You will see the problem. >> 4) Stop and start this instance and the problem goes away. >> >> Our production environment relies on automatically launched and loaded >> instances via EC2 api. >> >> Has anyone with knowledge of this particular ArangoDB AWS image an idea >> about what why it happens? >> >> >> Thanks in advance >> Soren >> >> -- You received this message because you are subscribed to the Google Groups "ArangoDB" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
