f3fec9a33c
Handle 402 response on app installation
2017-06-19 12:17:55 +02:00
40aaffe365
tests: Fix usage of settings.setDnsConfig
2017-06-15 20:05:35 -07:00
3745e96a6f
domain -> fqdn
2017-06-15 19:56:04 -07:00
9ead482dc6
Make verifyDnsConfig take zone name
...
part of #110
2017-06-15 19:55:24 -07:00
865c0a7aa7
Pass other level domains to dns API backends
...
part of #110
2017-06-15 19:55:01 -07:00
c760c42f92
make waitForDns take zone name argument
...
part of #110
2017-06-15 19:54:08 -07:00
ded31b977e
Add config.setFqdn and config.setZoneName
...
Part of #110
2017-06-15 19:53:20 -07:00
4781c4e364
Deliver empty JSON object on success
...
This ensures the client does not throw a parsing exception
2017-06-15 07:49:19 -07:00
658cbcdab9
bump mail container version (catchall support)
...
part of #33
2017-06-15 07:48:57 -07:00
0cc980f539
Add setting for catch all address
...
Note that this is not a flag on the mailboxes because we might theoretically
support forwarding to some other external domain in the future.
Part of #33
2017-06-15 07:48:46 -07:00
87ca05281d
Revert "Always check for updates prior to performing an update"
...
Lets keep the rest apis more single purpose and offload this case to the
client
This reverts commit 0bddd5a2c6
2017-06-13 22:58:07 +02:00
0bddd5a2c6
Always check for updates prior to performing an update
...
This covers the case where the box has not yet received a tarballUrl but
the user already setup a subscription.
2017-06-13 21:42:32 +02:00
20f2a6e4c6
Block updates if sourceTarballUrl is missing
2017-06-13 21:33:03 +02:00
6d47737de7
Remove unused require
2017-06-13 21:14:27 +02:00
23532eafea
Fix path to version
2017-06-07 20:31:18 -07:00
c0c07c2839
ensure .ssh dir exists
...
Fixes #349
2017-06-07 09:50:31 -07:00
daa4c66e7f
Do not perform automatic updates for major platform version
2017-06-05 18:06:00 +02:00
414e9bdf05
Do not use lastBackupId in cleanup logic
...
lastBackupId is only used as a "message" passing field for apptask restore.
Theoretically, this code somehow protects a race between the cleanup logic
and the restore apptask. this is unlikely to happen and adds unnecessary
complexity.
2017-06-01 14:47:57 -07:00
c342e52e7d
Record copyLastBackup in the backupdb
2017-06-01 14:08:55 -07:00
78aa9c66f7
Add a note why we do not cleanup more aggressively yet
2017-06-01 10:33:49 -07:00
986ec02ac6
Add debug on what backup is preserved
2017-06-01 09:38:39 -07:00
4e0bb9187a
lower case domain in migrate code path
2017-06-01 09:26:03 -07:00
9c8a8571b4
Ensure we lowercase the domain name before consuming it in dns setup
...
Finally fixes #335
2017-06-01 17:29:46 +02:00
0c9dce0c9f
redis: set memoryLimit to 600 because only half is RAM
2017-05-31 23:09:47 -07:00
471539d64b
CNAME output from dig has trailing dot
2017-05-30 21:14:28 -07:00
f34d429052
kill the backup process if it runs for too long
2017-05-30 16:11:12 -07:00
82e53bce36
ensure backups and clean them every 6 hours
...
also, make sure they don't run at the same time.
2017-05-30 16:04:32 -07:00
b04a417cfc
Cleanup errored and creating backups
...
Fixes #330
2017-05-30 15:16:08 -07:00
77641f4b51
Add backupdb.getByState and backupdb.getByTypeAndState
...
part of #330
2017-05-30 14:30:06 -07:00
765d20c8be
Add backup states to track unfinished backups
...
part of #330
2017-05-30 13:43:30 -07:00
d2420de594
refactor backup cleanup logic
2017-05-30 13:43:30 -07:00
ddb69eb25c
remove native-dns and use dig directly
...
native-dns module is unmaintained and we keep getting sporadic
errors from that module
Fixes #220
2017-05-26 16:51:05 -07:00
11697f11cf
use constants for admin location
2017-05-24 15:41:37 -07:00
6fc69c05ca
Add noop storage backend
...
This is sometimes useful when an update gets stuck because of some
bug in backup logic.
Note that you cannot restore from this backend because nothing is
saved.
2017-05-22 10:45:01 -07:00
65cff35be6
Do not dump certs in the log files
2017-05-19 14:39:08 -07:00
7467907c09
Do not dump data in update script since it might have the cert
2017-05-19 14:34:20 -07:00
d6c32a2632
tweak redis memory limit based on app's memory
2017-05-18 15:39:38 -07:00
7dc277a80c
Give the backup task more memory 150 is often too close to the limit
2017-05-17 14:17:54 +02:00
88e844b545
Bump infra version to reconfigure apps for http2 support
2017-05-12 16:25:14 -07:00
d164f881ca
Bring back code for alt domain match
...
There are no actual tests for this yet. Should be added.
2017-05-11 21:55:29 +02:00
4994a5da49
Use -checkhost openssl subcommand
2017-05-11 21:31:01 +02:00
374130d12a
Only set local dns server if run on a cloudron
2017-05-11 15:37:44 +02:00
05fcdb0a67
Extract CN from cert with JS
...
unlike the sed script, this does not rely on the order openssl reports the subject entities
2017-05-11 15:19:02 +02:00
23827974d8
Fix certificate validation to work with new openssl version as well
2017-05-11 14:58:29 +02:00
ae2c0f3503
Use new mail container (fix for exec)
2017-05-10 21:58:39 -07:00
cbb93ef7ad
For low end cloudrons, give a delay between addon starts
...
Starting them all at once, sometimes hogs cpu/memory too much
and makes the startup scripts of the addons error.
The new addons setup a .setup file to confirm initialization.
In a future commit, we can use those .setup files to check if
the addon has started up instead of a timeout
2017-05-10 15:43:02 -07:00
4d3c6f7caa
better error message
2017-05-09 11:24:47 -07:00
6ef2f974ae
fs: Use key to determine backup extension
2017-05-08 16:03:29 -07:00
ffc0a75545
user.get() returns UserErrors
2017-05-08 13:51:19 +02:00
8ae65661dd
redact the password so it is never displayed in logs
2017-05-05 15:36:47 -07:00
Johannes Zellner