e98a1a9767
docker container can use system dns
...
only mail container needs unbound for dnsbl
2024-04-29 15:48:30 +02:00
203330d1b8
lint: const
2024-04-29 13:05:07 +02:00
c8d66384c7
domains: check if wildcard is boolean in tlsConfig
2024-04-29 12:52:12 +02:00
74447d2690
lint
2024-04-29 12:49:20 +02:00
b66ddedc86
domains: remove unused wildcard check
2024-04-29 12:45:57 +02:00
cd5cae33ce
dns: switch over to systemd for the host
...
this changes unbound to listen to 127.0.0.150 (150 is roman CL)
we cannot only bind on docker bridge because unbound is relied
upon for the initial domain setup. docker itself is only initialized
when the platform initializes
2024-04-29 11:06:03 +02:00
7eda1136ea
oidc: starting with new .json model files is not worth a log line
2024-04-29 09:19:37 +02:00
efa1acddd4
dns: unregister domains if type is disabled
2024-04-27 18:43:31 +02:00
e00db115ad
restore: fix crashes
2024-04-27 12:46:37 +02:00
366f247910
oidc: only start in set callback
2024-04-27 11:51:10 +02:00
2a6368af60
remove usage of constants.DASHBOARD_SUBDOMAIN
2024-04-27 11:10:24 +02:00
5420630453
oidc: start the server when dashboard domain is set
...
the activation logic has changed to use oidc flow. this requires
the oidc server to be started and available. otherwise, the redirection
after owner creation fails.
2024-04-27 11:02:50 +02:00
4e39eb89fd
const
2024-04-27 10:48:23 +02:00
8a987db177
provision: add route to detect ipv4 and ipv6
2024-04-26 20:53:32 +02:00
126587ba82
lint: constness
2024-04-26 20:09:36 +02:00
860ebcbe6a
provision: add activation guard
2024-04-26 20:06:56 +02:00
2da361a1f2
waitfordns: resolve and check against NS' IPv6 address
2024-04-26 19:12:53 +02:00
23e20b9b83
waitfordns: better debugs
2024-04-26 14:46:01 +02:00
2f425f8119
provision: add ipv6 config
2024-04-26 12:20:15 +02:00
f6e4f1aefc
network: ipv4 can be disabled
2024-04-25 15:50:42 +02:00
cae2bfbdc2
domains: add desec provider
2024-04-24 21:29:42 +02:00
3787f90283
appstore: bump timeout to 60s instead of 30s
...
this timeout is hit on some servers (which have some networking
issue). unfortunately, this triggers a bug in superagent -
https://github.com/ladjs/superagent/issues/1801
2024-04-23 11:41:51 +02:00
c2326bc5cc
oidc: add rate limit for login requests
2024-04-21 20:58:12 +02:00
ec7dabc1c7
oidc: also allow login on aliased app domains
2024-04-19 19:03:23 +02:00
d137cdf881
update cron module
...
CronJob -> CronJob.from
CronJob(time) -> CronTime
2024-04-19 18:31:47 +02:00
4516b0c57c
Do not return but continue in a loop
2024-04-19 14:29:41 +02:00
49243822af
dashboard: show app checklist
2024-04-19 14:17:54 +02:00
16521d5434
Fix updateChecklist usage
2024-04-19 12:15:13 +02:00
1afa2e87ec
mailserver: a056bcfd broke mail server restart
...
after proxying, we never restarted the mail server
also add note that restart has to reconfigure
2024-04-19 10:48:08 +02:00
7d6636bb54
Only add checklist items if they apply due to sso state
2024-04-18 16:05:38 +02:00
3c7e6b59f0
Add initial support for apps.checklist
2024-04-17 16:54:54 +02:00
daa8a60da2
oidc: Inject currently hardcoded CLOUDRON_OIDC_PROVIDER_NAME env var
...
This is designed to be used in the packages for the login button:
"Login with ${CLOUDRON_OIDC_PROVIDER_NAME}"
2024-04-17 15:06:22 +02:00
f231d51d0b
Make oidc authproxy login button translatable
2024-04-17 14:21:07 +02:00
a572374ad7
updatechecker: deep compare update object from appstore
...
When 'changelog' , 'unstable' fields change the box code is not
getting it.
2024-04-16 19:30:14 +02:00
1cf315634c
appstore: check the type of unstable field
2024-04-16 19:19:27 +02:00
b0d2bdbad9
Make it login with cloudron in authproxy
2024-04-16 14:56:18 +02:00
255fb0cac0
proxyauth: show intermediate login button page
2024-04-16 13:43:12 +02:00
c3be0018fe
proxyauth: send user to oidc login instead of /login
2024-04-16 11:29:00 +02:00
1008ec4fa1
proxyauth: remove basic auth login form
2024-04-15 18:52:07 +02:00
21d7438bbe
proxyauth: user OpenID instead of basic auth
2024-04-15 15:59:16 +02:00
0a748ac78a
better AVX error message
2024-04-15 10:10:13 +02:00
76c4002a04
oidc: Add profile picture claim
2024-04-14 12:05:45 +02:00
36ab5800a3
oidc: enable CORS for internal apps
2024-04-11 19:10:29 +02:00
f11becfcc8
async'ify
...
crazy this has gone unnoticed for so long!
2024-04-10 18:52:39 +02:00
87ae95aa4f
Add per-app notes feature
2024-04-10 18:34:58 +02:00
b4e4f26361
Rework cpuShares into cpuQuota
...
cpuShares is the relative weight wrt other apps. This is used when
there is contention for CPU. If we want this, maybe we implement
a UI where we show all the apps and let the user re-order them.
As it stands, it is confusing.
cpuQuota is a more straightforward "hard limit" of the CPU% that you
want the app to consume.
Can be tested with : stress -c 8 -t 20s
2024-04-10 18:25:14 +02:00
efd0be5e2c
services: send the default memory limit
2024-04-10 12:42:25 +02:00
8b7dca00af
app memory: make slider go till RAM
...
anything above RAM is useless
2024-04-10 12:12:49 +02:00
be2775e12e
memoryLimit: redefine to not include swap
...
Currently, we allocate 50% as RAM and 50% as swap. The manifest is
usually quite conservative on memory values. This means that we set
up a system where the app is applying memory pressure almost immediately.
This then swaps things randomly and increases cpu usage (kswapd shows
up in the profile).
To rethink the whole situation: we should not cap apps with a swap limit at all.
The memory hard limit is what is important. By redefining memoryLimit , we are
doubling every container's memory and it's good that we over allocate this.
2024-04-09 18:59:40 +02:00
2b1b304c6e
backup/import/restore: fix crash with root path calcuation
...
rootPath was calculated before the arguments were validated
2024-04-09 13:53:48 +02:00
Girish Ramakrishnan