a year ago
My mongo server was running fine for weeks but suddenly started crashing and I can't get it back online to either serve my web app or extract the data to spin up a new container.
I initially thought it might be a resources issue so I subscribed to the Hobby plan and increased the resources, but that hasn't helped.
I have a mongodump of some of the data from my server but I would lose a fair bit if I wrote off this volume and started afresh.
Looking at the log, it's not clear what exactly is the cause but I can't get it online for long enough to do anything with it.
```
container event container restart
Automatically disabling TLS 1.0, to force-enable TLS 1.0 specify --sslDisabledProtocols 'none'
Initialized wire specification
Implicit TCP FastOpen unavailable. If TCP FastOpen is required, set tcpFastOpenServer, tcpFastOpenClient, and tcpFastOpenQueueSize.
Successfully registered PrimaryOnlyService
Successfully registered PrimaryOnlyService
Multi threading initialized
Starting TenantMigrationAccessBlockerRegistry
MongoDB starting
Memory available to mongo process is less than total system memory
Build Info
Operating System
Options set by command line
Detected unclean shutdown - Lock file is not empty
Storage engine to use detected by data files
Recovering data from the last clean checkpoint.
Using the XFS filesystem is strongly recommended with the WiredTiger storage engine. See http://dochub.mongodb.org/core/prodnotes-filesystem
Opening WiredTiger
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger message
WiredTiger opened
WiredTiger recoveryTimestamp
/sys/kernel/mm/transparenthugepage/enabled is 'always'. We suggest setting it to 'never' in this binary version vm.maxmap_count is too low
Updated wire specification
current featureCompatibilityVersion value
Clearing temp directory
Initializing cluster server parameters from disk
Flow Control is enabled on this deployment
Initializing full-time diagnostic data capture
Setting new configuration state
Timestamp monitor starting
Listening on
Listening on
Listening on
Waiting for connections
mongod startup complete
Unclean full-time diagnostic data capture shutdown detected, found interim file, some metrics may have been lost
Writing fatal message
Writing fatal message
Writing fatal message
Frame
Frame
BACKTRACE
Frame
Frame
Frame
Frame
Frame
Frame
Frame
Frame
Frame
Frame
Frame
Frame
Frame
container event container died
ⓘ Deployment information is only viewable by project members and Railway employees.
5 Replies
a year ago
You're on the trial plan and are limited to 500MB Volumes, MongoDB can not happily run with only 500MB, you will need to upgrade to the Hobby plan for access to 5GB volumes, after the upgrade, go into the volumes settings to grow it to 5GB.
a year ago
I updated to Hobby in my attempts to resolve this before opening the thread. Mongo still can't boot with 8GB assigned
Attachments
a year ago
I did not realise that the volume was independently clickable from the service
I had reviewed https://docs.railway.app/overview/the-basics#volume-settings from your first message, but thought that the view for Postgres was just different from Mongo rather than that the bottom half of the card was clickable separately.
That has gotten it back online.
Attachments
Status changed to Solved brody • over 1 year ago