FAA Solution to Missing dependencies from USM?(2)

[francois]

@brokenman:
This has gone quite a bit off topic. Can someone confirm that the OP's original query has been satisfied? I will answer the other questions briefly here but Francois please do start a relevant thread.

There is no intent in my part to provide additional burden to the developpers, especially that they went for a system of moderators to free them to be able to work on other tasks. There is no hurry for me.

For additional discussions about the fine tuning of porteus usm package manager you are invited to join the following thread. Fine tuning of usm for porteus use:
http://forum.porteus.org/viewtopic.php?f=48&t=4552

[brokenman]

every 6 hours the script on the server downloads latest time stamps from every repo changelog and compares them with latest saved.
if there is a mismatch then updates database, if not then skips.

This is essentially what the update script does. The load comes because if one package has been updated (for example in slacky) then the server needs to process every package in the (slacky) repo causing server strain. If this is done every 6 hours then it could be that every 6 hours the server is strained. This is why I have chosen every 3 days. If I can get LXC containers running then strain wouldn't be a problem as processor restrictions can be set in place. The current system doesn't support this.

While I am in here, today I added the default MODULES=true so that by default porteus modules will be generated on Porteus. I see the sense in what Francois said about a package manager actually installing the packages by default. I also added a general slackware package update function whereby all the latest patch files from slackware can be downloaded and installed. There are certain caveats using this in Porteus since the actual package resides inside a base module, but the function is now there for people that want it (on github only at the moment) in the CLI end only as a testing ground.

[Ed_P]

If this is done every 6 hours then it could be that every 6 hours the server is strained.

But what are the odds of an update every 6 hrs? Is there any analysis of the dbs update frequencies the past few months? Maybe every 8 hrs or 12 hrs would work better?

[brokenman]

The logs are saved so I could go through them manually and see how often they are updated. I might be less time consuming for me to just set up a cron job to create a running log with stats on how often they are updated and check it in a week. In any case I would like to setup LXC on the server so this problem is history. This means I have to learn all about how the containers work which is also a time muncher.

Hamza, what is you opinion on the server strain and if we can afford to straint he server for 15-25 minutes every day? If once a day is not a problem then we can run the 32 bit side one day and the 64bit side the next.

[Ed_P]

I might be less time consuming for me to just set up a cron job to create a running log with stats on how often they are updated and check it in a week.

I'm not sure what a cron job is but I like the sounds of that approach.

[Bogomips]

This means I have to learn all about how the containers work which is also a time muncher.

After all this is a community effort, so someone more knowledgeable in this field can gen up on lxc containers and help brokenman out? Is there anyone out there?

[beny]

this is lxc check i have to rebuild the kernel for missing option i see:
bash-4.3# lxc-checkconfig
--- Namespaces ---
Namespaces: enabled
Utsname namespace: enabled
Ipc namespace: enabled
Pid namespace: enabled
User namespace: enabled
Network namespace: enabled
Multiple /dev/pts instances: enabled

--- Control groups ---
Cgroup: enabled
Cgroup clone_children flag: enabled
Cgroup device: enabled
Cgroup sched: missing
Cgroup cpu account: missing
Cgroup memory controller: enabled
Cgroup cpuset: enabled

--- Misc ---
Veth pair device: enabled
Macvlan: enabled
Vlan: enabled
File capabilities: enabled

Note : Before booting a new kernel, you can check its configuration
usage : CONFIG=/path/to/config /usr/bin/lxc-checkconfig
i try to compile only the missing line work in progress,i can't rebuild because i have the zen patch if i have this patch the missing line are unavailable...sorry

[Hamza]

I'm working on the issue.

[brokenman]

Thanks Hamza.

Beny the container won't be run on Porteus. It will be run server side.