=== xavpaice- is now known as xavpaice | ||
flxfoo | hi all | 09:41 |
---|---|---|
stickupkid | flxfoo, hi | 09:43 |
flxfoo | I have a little using python script, which now returns those lines: RPC: Connection closed, reconnecting | 10:00 |
flxfoo | twice | 10:00 |
flxfoo | then | 10:00 |
flxfoo | Receiver: Connection closed, reconnecting | 10:00 |
flxfoo | RPC: Automatic reconnect failed | 10:00 |
flxfoo | nothing as changed on that machine, last thursday script was used successfully... I tried to uninstall/install (pip3 install juju) again (not my favorite thing) no luck | 10:03 |
flxfoo | actually not, the script worked between the saturday-sunday night | 10:03 |
flxfoo | last night it failed | 10:04 |
flxfoo | any input how I can try to debug that? | 10:04 |
flxfoo | in another hand, can I use like "curl" to retrieve what I want? if yes, any documentation? | 10:29 |
stickupkid | flxfoo, can you see what juju debug-log says, it might be able to give us some missing information | 10:31 |
flxfoo | it is from an instances which is not a juju node | 10:38 |
flxfoo | well I will check if there is something | 10:39 |
flxfoo | nothing in debug-log | 10:40 |
flxfoo | stickupkid:do you know how could I get node informations using a curl request for example? | 10:42 |
stickupkid | flxfoo, what information are you wanting to get? | 10:48 |
stickupkid | flxfoo, what version of juju are you using, and I'm assuming you're using pylibjuju? | 10:49 |
flxfoo | stickupkid:guess so, installed via `pip3 install juju` | 10:54 |
flxfoo | version is 2.8.3 | 10:54 |
stickupkid | flxfoo, yeah, that's pylibjuju | 10:54 |
stickupkid | flxfoo, what are you calling in juju? | 10:54 |
flxfoo | Successfully installed bcrypt-3.2.0 certifi-2020.6.20 cffi-1.14.2 chardet-3.0.4 cryptography-3.1 idna-2.10 juju-2.8.3 jujubundlelib-0.5.6 macaroonbakery-1.3.1 mypy-extensions-0.4.3 paramiko-2.7.2 protobuf-3.13.0 pyRFC3339-1.1 pyasn1-0.4.8 pycparser-2.20 pymacaroons-0.13.0 pynacl-1.4.0 pytz-2020.1 pyyaml-5.3.1 requests-2.24.0 setuptools-50.3.0 six-1.15.0 theblues-0.5.2 toposort-1.5 | 10:54 |
flxfoo | typing-extensions-3.7.4.3 typing-inspect-0.6.0 urllib3-1.25.10 websockets-7.0 | 10:54 |
flxfoo | I getting public ips of instances | 10:55 |
flxfoo | the main lines related to connection, would be creating a controller, await controller.connect(), and await model.get_status() | 10:56 |
flxfoo | juju controller version is 2.8.0 | 10:57 |
flxfoo | I tried to force a version `pip3 install juju==2.8.0` which actually install a `2.8.1` version, don't ask me why... :) | 11:06 |
flxfoo | connection is done like the following example: https://pythonlibjuju.readthedocs.io/en/latest/narrative/controller.html#connecting-with-an-explicit-endpoint | 11:23 |
flxfoo | stickupkid: `pip3 install juju=2.7.1` actually install juju v2.8.0... still same error | 11:30 |
flxfoo | stickupkid:audit.log in controller seem to have some things: {"errors":{"conversation-id":"4dbb8fc47f88a5bd","connection-id":"9D","request-id":12,"when":"2020-09-14T11:32:58Z","errors":null}}" | 11:33 |
flxfoo | stickupkid: machine-0.log (which is the controller itself according to the index number) | 11:35 |
flxfoo | I have this: | 11:35 |
flxfoo | 2020-09-14 11:34:19 ERROR juju.rpc server.go:600 error writing response: *net.OpError write tcp 172.31.29.252:17070->172.31.32.214:54856: write: connection reset by peer | 11:35 |
flxfoo | 2020-09-14 11:34:19 ERROR juju.apiserver apiserver.go:939 error serving RPCs: codec.ReadHeader error: error receiving message: websocket: close 1009 (message too biY | 11:35 |
flxfoo | so the websocket is closed because of this "message too big" | 11:36 |
stickupkid | flxfoo, that's strange, never seen that before | 11:41 |
stickupkid | flxfoo, do you mind opening a post on discourse, so more eyes can see it | 11:41 |
flxfoo | sure | 11:41 |
flxfoo | It could be a "hardware" problem... after running pip3 install, that terminates with a seg fault | 11:42 |
flxfoo | ok, I will try to create a new vm and see if that appears... that sounds like a crappy vm, or something I will let you know.... If I can reproduce I will open a post. | 11:43 |
flxfoo | stickupkid:thanks for your support. | 11:43 |
flxfoo | stickupkid:new ubuntu server (just went from bionic to focal) , no more segfault on pip install, installing same libpyjuju as the controller version, 2.8.0, same RPC error , same message in controller machine-0.log. | 12:18 |
flxfoo | stickupkid:tried to play with `max-frame-size` but no luck=same error | 12:18 |
flxfoo | stickupkid:I have a repeated eror line like so `ERROR juju.worker.dependency engine.go:671 "api-caller" manifold worker returned unexpected error: [619092] "machine-0" cannot open api: model cache: model "61909253-939f-48c0-8452-389111410e43" did not appear in cache timeout` | 12:20 |
flxfoo | does that ring bell? | 12:20 |
flxfoo | stickupkid_:can I retrieve information or run actions using some curl magic? | 13:28 |
stickupkid_ | flxfoo, this is concerning ` model cache: model "61909253-939f-48c0-8452-389111410e43" did not appear in cache timeout` | 13:56 |
hml | stickupkid_: I can’t approve my own pr :-) 11967 looks good, a added a few comments and a nit | 15:16 |
stickupkid_ | ta | 15:21 |
stickupkid_ | hml, ho? | 15:28 |
stickupkid_ | or are you in a meeting? | 15:28 |
stickupkid_ | if so catch after that | 15:28 |
hml | stickupkid_: meet you in daily | 15:28 |
Moore | deployed my first cluster with juju but kubectl not connecting | 17:02 |
Moore | nickmoore@KubeMasterNM:~$ kubectl get pods -n kube-system | 17:02 |
Moore | is there anyone available to answer a quick question on an azure deployment? | 17:21 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!