Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
✪ vi4m Marcin Kliks
@vi4m
@sundaresanm we don't support competitors systems ;-)
damian1976
@damian1976
I think you should ask authors of the tool you are using, not ralph guys.
✪ vi4m Marcin Kliks
@vi4m
but you always can migrate your data to the ralph system using csv
aveseb
@aveseb
@vi4m I saw that your were talking about ralph_scanner, is this step is launched ?
✪ vi4m Marcin Kliks
@vi4m
@aveseb yes
i've preprared some basic engine for it
aveseb
@aveseb
Cool :-)
is there a branch where we can find it ?
✪ vi4m Marcin Kliks
@vi4m
i have to polish it a bit, will push it somewhere there: https://github.com/vi4m/ralph_scanner
Brian Choate
@brianchoate
In 3.0, is there a trick to importing racks and having them match up with existing server rooms? I have tried with both the server room name and id. The rest of data in the CSV seems to load just fine.
✪ vi4m Marcin Kliks
@vi4m
well. you have _id columns which should be matched
Brian Choate
@brianchoate
I've tried using both server_room and server_room_id in the input csv. Neither seemed to get mapped when importing.
This is what the file looks like
server_room_id,id,name,description,orientation,max_u_height,visualization_col,visualization_row,server_room_name
12,,rack-a01 - Equinix Rack# 62230-0105,,1,42,0,0,ASH
8,,FRA1,,1,5,0,0,FRA
9,,rack-p1,,1,44,0,0,POZ
10,,rack-pozo1-office,,1,37,0,0,POZ Office
7,,RackSpace,,1,10,0,0,Cloud
damian1976
@damian1976

@brianchoate What I can see in 2.x, rack is exported in csv file (and then imported to 3.x) in th following format:

id,name,server_room,description,orientation,max_u_height,visualization_col,visualization_row

(I don't know if racks' ids can be empty/NULL or need to have predefined ids)

✪ vi4m Marcin Kliks
@vi4m
@brianchoate how's going with imports?
Brian Choate
@brianchoate
@damian1976 If the id is present it will update an existing record. If it is missing then it creates a new one and sets the id.
@vi4m I was not able to get the import to work with the server rooms links. I was able to import the records without the reference and then manually edit each one to set the server room. Not ideal, but got me past that point.
Now I am struggling to import server/hardware data.
Brian Choate
@brianchoate
When trying to import hardware I get this error: Line number: 1 - Cannot assign None: "DataCenterAsset.model" does not allow null values.
There is a model field in the data and it is not null.
✪ vi4m Marcin Kliks
@vi4m
@brianchoate i think i localized the bug
Brian Choate
@brianchoate
@vi4m so it is a bug in the code base? Will there be a patch for it?
✪ vi4m Marcin Kliks
@vi4m
yeah, the code handled ralph 2.0 csv compatible file, i added simple setting switch file to disable it
work being done here: allegro/ralph#2192
✪ vi4m Marcin Kliks
@vi4m
Zrzut ekranu 2016-01-30 o 21.27.12.png
new ralph3 start screen :)
PY
@fossilet
👍
damian1976
@damian1976
:) waiting for scrooge & scan :)
✪ vi4m Marcin Kliks
@vi4m
we deployed ralph 3.0 in 25% of our company already, but the remaining 75% of users needs deployment system
so this will be our next point on roadmap, including scrooge integration
Manuel Rüger
@mrueg
Are there any plans to release ralph again on pypi? Right now I'm working on deploying ralph, gunicorn, nginx with saltstack and it would be helpful to have it on pypi for that reason
✪ vi4m Marcin Kliks
@vi4m
@mrueg ralph 3?
why not deb packages then? rpm based distro?
✪ vi4m Marcin Kliks
@vi4m
technically speaking it should be easy to install it as pure-python library directly from github
trying to do this now, just to check ;)
pip install git+https://github.com/allegro/ralph@ng
but there are more steps to run after installation to collect javascript dependencies
install-js: npm install bower instal
in the feature we don't guarantee it will work this way, maybe some additional steps will be introduced
that's the reason we prefer to have docker, deb and other native packages formats
Manuel Rüger
@mrueg
@vi4m I tried the deb packages, but ralph3 tried to write a logfile into sitedir. That's why I moved back to a virtualenv
this path: /opt/ralph/ralph-core/lib/python3.4/site-packages/ralph/runtime.log
✪ vi4m Marcin Kliks
@vi4m
it's easy to fix it, then
✪ vi4m Marcin Kliks
@vi4m
allegro/ralph#2217
you can then set different LOG_FILEPATH directory from environment variable
Manuel Rüger
@mrueg
thanks :+1:
✪ vi4m Marcin Kliks
@vi4m
@mrueg new deb snapshot available with LOG_FILEPATH env variable parsing
Manuel Rüger
@mrueg
still trying to run it with gunicorn. https://github.com/allegro/ralph/blob/4191332617eab11153629df73e49c15df67d96a0/src/ralph/__init__.py#L12 this line causes an issue for me right now
File "/home/ralph/ralph-20160217.3/src/ralph/init.py", line 7, in monkey_options_init
self.old_init(meta, app_label)
File "/home/ralph/ralph-20160217.3/src/ralph/init.py", line 13, in <lambda>
self, meta, app_label
RuntimeError: maximum recursion depth exceeded