Add stats for pilot #90
Conversation
This needs folder /home/buendia to exist and contain file server_status.txt with write permissions for buendia scripts. Current implementation is for these to be created on setup if they don't exist by Yocto.
Patients weren't able to be synced until they were edited, due to a bug in the SQL trigger that creates entries in the `buendia_patient_sync_map` on person insert. This change fixes that bug. Fixes projectbuendia/client#187
Previously, secondary concepts in chart fields (any concepts other than the first listed concept) were not synced to the client. This resulted in the client not knowing how to render these secondary concepts unless they were also a primary concept in another chart field. Fixes projectbuendia/client#202.
Fix patients not syncing until an observation was added.
Sync secondary concepts that aren't referenced anywhere else.
- Allow free-text observations to be submitted as part of encounters. - Attribute encounters to the user who entered them. - Standardise the way observations are represented in encounters, so they more closely match the representation used for observation sync.
…-default-form-section Create default form section [invisible]
…-unknown In profile, make yes_no really mean yes or no and introduce yes_no_unknown
Server-side changes required to support notes panel.
…re/buendia-backup-status-reporting
…uired-fields Outputs the required property of fields.
…patient-like-search Enable like search.
…tus-reporting Feature/buendia backup status reporting
| @@ -42,7 +53,7 @@ Content-Type: text/html | |||
|
|
|||
| <h1>Project Buendia server</h1> | |||
|
|
|||
| <h2>Staff access</h2> | |||
| <h2>Bokoro Pilot - Server 9 (ATFC Simple Server)</h2> | |||
There was a problem hiding this comment.
This could have a TODO to make it dynamic and not hard coded.
|
My only concern is the shutdown request check interval. Everything else seems fine. |
|
Thanks @llvasconcellos. I've changed the interval for the shutdown and reboot services to 30s, and added a TODO as you suggested, also removing the specific server name which probably shouldn't be committed. |
|
The dashboard stuff looks good, but the shutdown checker and the reboot checker should be unified - currently we've got a service in Debian checking for shutdown and a service in Yocto checking for reboots. |
|
Also, the fact that so many different pieces of code have to be able to interpret status codes is pretty scary. |
|
@capnfabs - both checks are on the Yocto side I think. I did the shutdown check in the same way as the reboot check. Do you recommend merging them both into a single service that checks for both reboot and shutdown requests then? |
|
@dancunningham Ah yes, you're right - sorry about that. |
|
@capnfabs @llvasconcellos does anything need done to this before we can merge it in? I'd like to verify that it works on an Edison. For that we need to merge it and do a release anyway! |
|
@dancunningham It's fine by me. You can merge it. |
|
|
||
| buendia_status=$(cat /home/buendia/server_status.txt) | ||
|
|
||
| if [ $buendia_status -eq 2 ]; then |
There was a problem hiding this comment.
You can use bash case statements for this: http://tldp.org/LDP/Bash-Beginners-Guide/html/sect_07_03.html.
Also, as is, you should be quoting the $buendia_status variable in these tests because you're using [ as opposed to [[. See http://serverfault.com/a/52050/145418
|
All in all, looks good, but back over to you for adjustments and tweaks @dancunningham :) |
|
This PR has been reconstituted in #117. |
Adds some additional tools to the "dashboard" for the pilot team to use:
To deploy this on a live server, all the files need to be put in place in Debian and then the commands in /packages/buendia-utils/control/postinst need to be run.
Alternatively, you could build new versions of the buendia-dashboard and buendia-utils packages and let the autoupdate script install those from USB.