yyoung: might it be some javascript is blocked by your webbrowser?
yyoung
IIRC it was fine before and I didn't change anything since then, I'll have a look
yvanzo
yyoung: The script batch.js must be able to query getconfig to retrieve the extra buttons.
yyoung
yvanzo: It works fine under Chrome incognito mode, in normal mode I can still see a successful request to getconfig, but there's a "RPC: request rejected (bad origin): https://tickets.metabrainz.org"; message in the console
yvanzo
yyoung: might be caused by an add-on that is enabled in normal mode only, try disabling it if possible.
yyoung
yvanzo: Yes, I'll have a check, thank you
sampsyo has quit
sampsyo joined the channel
nelgin
OK, let me take a look
sumedh joined the channel
alastairp
hi _lucifer
cool, so we know that the email works, and we know that the dump works with the storage box
nelgin
yvanzo, vacuuming db is running now.
_lucifer
Hi alastairp!
yes, right.
alastairp
_lucifer: I'm just reading the dump file again. so it did the dump to the temp directory, and then copied it to the backup dir, then failed to copy it again to the ftp dir?
Lotheric has quit
ah yes, I see the incremental dump in the backups directory
I see the permissions on ftp directory being owned by root. that's interesting.
_lucifer
alastairp: yes. it failed while trying to create a directory under ftp.
alastairp
so yes, I think it makes sense to change that back, I'll run through the dump process again to check
_lucifer
right my guess is that this got mixed when lemmy went down and dumps were recovered.
alastairp
there are two additional changes we could make: 1) add some explicit checks at the beginning of the dump like we do for tmp location - check that the permissions of all dump locations are correct and that we can expect to continue
2) integrate sentry bash so that in case of an error we get an explicit notification instead of having to go back and look through the logs
nelgin
Thu 20 May 2021 06:53:36 AM UTC : UPDATE THE DB_SCHEMA_SEQUENCE IN DBDefs.pm TO 26 !
Very good.,
Can't locate List/AllUtils.pm in @INC (you may need to install the List::AllUtils module) (@INC contains: /musicbrainz-server/admin/../lib /etc/perl /usr/local/lib/x86_64-linux-gnu/perl/5.30.0 /usr/local/share/perl/5.30.0 /usr/lib/x86_64-linux-gnu/perl5/5.30 /usr/share/perl5 /usr/lib/x86_64-linux-gnu/perl/5.30 /usr/share/perl/5.30 /usr/local/lib/site_perl /usr/lib/x86_64-linux-gnu/perl-base)
at ./admin/BuildMaterializedTables line 10.
BEGIN failed--compilation aborted at ./admin/BuildMaterializedTables line 10.
Looks like a new dependency?
Looks like the docker is missing liblist-allutils-perl - How do I install it?
yvanzo that was trying to run "sudo docker-compose exec musicbrainz ./admin/BuildMaterializedTables --database=MAINTENANCE all" btw
not sure what chown does for a non-existent directory. so just rearranging the checks probably won't be enough.
yvanzo
(This isn’t a new dependency, rather an environment issue.)
revi has quit
revi joined the channel
alastairp
_lucifer: the problem is that this script runs as `lbdumps`, so if the ftp dir is owned by root, lbdumps can't do anything
which is interesting... maybe we could have a discussion to see if we run this as _root_, then we could make all of the necessary permission changes if needed. We can force the python scripts to run as a particular user if we want
let me put it on th elist
_lucifer
makes sense.
for now, we change the owner of the ftp manually ?
alastairp
yeah, I'll double-check that today
_lucifer
cool, thanks!
yvanzo has quit
MRiddickW has quit
reosarevok
ruaok, _lucifer (if you get support emails): see the latest email re: LB
yvanzo joined the channel
What happens if someone was to recreate their MB account with the same username? Would their LB account work again?
(if they have removed their MB account in the meantime)
_lucifer
no, I do not get support emails. i am not sure if deleting MB account also leads to deletion of LB account?
if no, then LB account would work again.
outsidecontext joined the channel
nelgin
yvanzo ok that seems to be working.
yvanzo has quit
adhawkins
Trying to update my container too, it's failing at step 22, trying to update yarn keys. Any thoughts?
So why won't it run inside docker? It's worked in the past...
Grrr...
yvanzo
it should eventually work inside the build
does it reliably work from the host, or maybe you were lucky?
do you use the same curl command from the host,
?
adhawkins
Yes, it works reliably on the host, and is now seemingly quicker.
I copied the command from the 'RUN' line echo'd during the build, and just replaced the NODE_VERSION variable with 1.16.0
yvanzo
16.1.0
adhawkins
Good spot. Just re-trying.
That works. It's slow again though.
loujine has quit
loujine joined the channel
loujine has quit
loujine joined the channel
When I say it works, I mean I can execute the curl command in the host.
yvanzo
adhawkins: two potential workarounds: either add options (--retry 10 --speed-limit 30 --speed-time 60 ) to that curl command, or copy the file you already download to build resources.
adhawkins
Already tried with retries 50, will add the speed ones too.
yvanzo
That is the main issue: Speed is too slow for you when downloading from nodesource.com.
adhawkins
Not sure what you mean about copying to build resources?
Speed switches didn't help.
yvanzo
This is definitely a load issue, your IP might not be redirected to the same nodesource server as we do.