GitLab

There is just too much variance across installed M2Crypto versions
and/or M2Crypto has too much internal churn to enable this by default,
sadly.  It just wasn't designed to expose this feature (the ability to
limit the time spent in SSL accept).

1. New table login_history to save all logins (uid,IP,portal).

2. Send warning email when user who has not been active for over a year
   logs in.

last run of assign, which is especially important on the moonshot
cluster which is multi-architecture.

I'm not sure if all containers will want stdin opened or attached on
console (i.e. for the master process), but for now we'll just turn it
on.

error gracefully instead of silent failure.

reservation graph.

I lifted a bit of code from elsewhere in the library, to compute the
Y value from the click event and map it to the line (dot) clicked on.
Pass that along to resgraph so that it can compute the type, and
pass that along to the reserve code.

do that, always intended to.

experiment permission, just membership. This solves the problem of users
having "user" privs in the project, but local_root in a subgroup. Users
still will not be able to reserve nodes.

This means that users can still use our ssh urls to reach containers
that don't run sshd.  We run a private sshd that has Ports to listen
on, and Match blocks containing ForceCommand directives, which run
docker exec $vnode_id <shell>.  User can configure which shell.

to warn user to use their reservation or die.

busier times. Add explanatory text to help modal.

Turns out the Linux raid tool (mdadm) will pick up fragmentary RAID
metadata and complain mightily.

work, but tested here okay. Who knows, does not affect us.

This reverts commit 0e4b83ef.

it does not hang up.