Hello Dan, I recently tried setting up a 4 node cluster of FOSSology 3.5.0 on Debian 9 machines and got it working with few tweaks.
I followed the guide written by Matt Taggart at https://github.com/fossology/fossology/blob/master/debian/fossology-common.README.Debian#L19 You can follow the same steps to do the installation. Then there are few changes required to make it working properly: 1. You need to NFS your report and logs directories as well otherwise while downloading the report, you'll get an empty page. 2. You will need to increase the maximum number of sessions per user by editing /etc/security/limits.conf and SSH config. * Since scheduler connects all the agent at startup (one connection per agent) which will easily exhaust the default settings in the system. I am planning to write a new wiki for the multi-system cluster. Thanks and Regards, Gaurav Mishra ________________________________ From: fossology@lists.fossology.org <fossology@lists.fossology.org> on behalf of Michael C. Jaeger <m...@mcj.de> Sent: 07 August 2019 01:08 To: Dan Stangel <dan.stan...@hpe.com> Cc: fossol...@fossology.org <fossol...@fossology.org> Subject: Re: [FOSSology] Anyone running FOSSology 3.x on a multi-system cluster? Hi, sorry to read this, I am not aware of anyone using the cluster setup. However, I was also wondering if it would be running. Not much has been changed since years however: https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Ffossology%2Ffossology%2Fcommits%2Fmaster%2Fsrc%2Fscheduler%2Fagent&data=02%7C01%7Cmishra.gaurav%40siemens.com%7Cb56d62a1a26447da0d6908d71aa5a4e1%7C38ae3bcd95794fd4addab42e1495d55a%7C1%7C0%7C637007171050837063&sdata=rBDRvaw4kHvtdplGNpxf0bAv%2BbOodgkARzsWbb8Arvw%3D&reserved=0 one thing were modifications to the max agents setting. kind regards, Michael > On 6. Aug 2019, at 18:50, Dan Stangel <dan.stan...@hpe.com> wrote: > > Hi there, > > Is anyone running FOSSology version 3.0 or higher on a multi-system > cluster? I am attempting to set up a new 3.5.0 in a cluster with one > scheduler/database system and two agent nodes, where the 3 nodes have > the repository NFS cross-mounted between them. > > I have run into many problems getting this working. It seems the > scheduler does not like the response it gets back from the remote > copyright and ecc agents - I may have a workaround for this. But with > other agents enabled (in /etc/fossology/mods-enabled) the scheduler > crashes with a seg fault, and no indication of what has gone wrong. > > Just curious if I'm the only one who has attempted this in recent > history, or if there are others out there with a healthy, happy 3.x > multi-system install working? > > Thanks, > Dan > > > > -=-=-=-=-=-=-=-=-=-=-=- Links: You receive all messages sent to this group. View/Reply Online (#3277): https://lists.fossology.org/g/fossology/message/3277 Mute This Topic: https://lists.fossology.org/mt/32746251/21656 Group Owner: fossology+ow...@lists.fossology.org Unsubscribe: https://lists.fossology.org/g/fossology/unsub [arch...@mail-archive.com] -=-=-=-=-=-=-=-=-=-=-=-
