bongfrog: haproxy is correct. all is working except the monster-ui

able to log in to monster-ui with superadmin. but shows 'too many requests' error when attempting to retrieve accounts or numbers

api requests work i.e. x.x.x.x:8000/v2/accounts/....../children does show children accounts

kazoo log when trying to get accounts via monster-ui shows

PUT: /v2/user_auth? from x.x.x.x

PUT request fulfilled in 80 ms 435K816B mem 43K327B bin

over and over and over.....

i get this after a while in the log: bucket x.x.x.x/no_account does not have enough tokens(1 needed) for this request

maybe your superduper admin password is incorrect?

monster-ui accepts the password for login and API requests accept the credentials

errors occur only in monster-ui when trying to retrieve accounts and numbers. navigating through other features (users, devices, voicemails, etc.) works

mc_ Any news when v5 will be released? It is October already...

btracht: I would take a look and see if there are any clues in the dev tools on the browswer you are trying to access monster from. Might be ints in the responces of the user auth requests

bongfrog: thanks. dev tools shows the request going to the original domain

bongfrog: any idea how to correct this?

again, it only happens with accounts and numbers

btracht: That is a head scratcher, I would go through all of the doc in system_config and see if anything pops out there with the old Domain

if monster is sending req to original domain, check the monster config.js (or whatever its called)

and update the api_url

then clear browser cache and reload monster

mentax: am asking around to see if there's a more firm date

I think you will also need to run: sup crossbar_maintenance init_apps /var/www/html/monster-ui/apps ${IP_ADDR}:8000/v2

mc_ BadRequest: i defined the api url in config.js and used sup command, did full reboot still same issue

...only with numbers and accounts

btracht: Just to make sure, the {IP_ADDR} in there is asking for the IP address of the server where Kazoo (specifically) Crossbar API is running.

correct

this is on a AIO server

AIO? Do you mean the DB also lives within this server?

yes

So, you moved the whole AIO server somewhere else?

Under a different domain?

If so, I would suggest you to, first backup your /etc/kazoo/ folder and then follow this guide and perform the appropriate changes whenever you see "IP_ADDR" and/or "_HOSTNAME" being used within a command.

installed and configured kazoo applications, fs, couchdb on new server (new domain). just copied actual DBs

I see.

What error(s) do you see on the logs when you try to log in with an account besides master account.

?*

Did you create this master account on the new installation or are you using the same you had before moving the DB to the new server?

using the original superadmin account. again, i only get the error when trying to access child accounts and numbers in monster-ui. i can get the child accounts and numbers of child accountsusing API requests

So, you can access any root account but not sub-accounts?

in monster-ui i can only get to account that i log into directly. clicking the dropdown tree or account app, just hangs and gives error

sounds like you need to compare your succeeding API req to the browser's attempt and see where they differ

the domain is wrong when attempting to get accounts. i have checked system_config caccount & rossbar docs. they all have the correct pvt_node. i'm thinking its a doc that i have not checked

but don't know which one

i have found a handful of docs in system_config with the original pvt_node that i have since changed. is there a way (sup command) i can sync the entire system_config DB with the updated domain?

not sure that that is even the issue, just something i noticed

bloody hell - activity.. i ask questions and there's no one for days lol

barney: lol

btracht: sup kapps_config flush

I think that commands takes care of flushing the old data out of the cache, so next time Kazoo tries to get that data, it will get the latest data from the config.

i think i tried that but i'll try it again

If it is outside system_config, you may need to use a more "aggressive" command but I guess it won't hurt on your cluster because it is not being used right now.

If that does not work for flushing the old information out of the cache, try running: sup kz_datamgr flush_cache_docs

no luck

still sending request to old domain

Oh, I think you need to reload Apache service after you modify config.js file for Monster-UI for the changes to take into effect.

sudo systemctl reload httpd.service

did that too

Not necessary to restart web server after config.js changes.