a month ago
Hi Railway team,
I'm experiencing a critical issue: my WordPress volume's disk has not physically resized despite the config being updated to 5000 MB and a successful deployment.
Impact: WordPress uploads fail with "The uploaded file could not be moved to wp-content/uploads/2026/08" — the volume is full.
Project details:
- Project ID: fda4524e-6f53-49b7-9a9e-fbee35f8a9f4 (impartial-acceptance)
-
- Service ID: 92483de9-0ac2-4438-8c97-ada622e43327 (souss-actualites)
-
- Volume ID: db68e6e5-ac5f-458b-b052-6811640db901 (souss-actualites-volume)
-
- Environment: production (28200222-6216-4488-8cd5-3c2a6b772855)
-
- Plan: Hobby
What happened:
- Volume config was originally 500 MB. Volume filled up, uploads started failing.
- I updated sizeMB to 5000 MB (Hobby plan limit) via the Railway canvas UI and clicked "Deploy Changes".
- Latest deployment a2b58e5c-46cd-4699-9ba3-6597fdfca99b completed with status SUCCESS (2026-08-31T13:02:54Z).
- VERIFIED via readContainerFileTool: /sys/block/zd512/size still returns 983040 sectors (~480 MiB). The host zvol was NOT resized.
Verified state (post-deploy):
- Volume config API: 5000 MB ✅
-
- Deployment status: SUCCESS ✅
-
- Actual host disk (/sys/block/zd512/size): 983040 sectors = ~480 MiB ❌ (expected ~10,485,760 sectors for 5 GB)
-
- WordPress uploads: still failing due to full disk ❌
Request: Please manually trigger the Live Resize for volume db68e6e5-ac5f-458b-b052-6811640db901 to actually expand the host zvol to 5000 MB. The config change exists but the physical resize automation did not fire.
Thank you!
1 Replies
Status changed to Awaiting Railway Response Railway • about 1 month ago
a month ago
RESOLVED — no staff intervention needed.
The Live Resize propagated to the host zvol a few minutes after the successful deployment. No manual action from Railway was required.
Final verified state (SSH + container):
- Host disk: 4.4 GB total, 4.0 GB free (was ~480 MiB before)
-
- Real write test: 20 MB file created instantly in wp-content/uploads/2026/08 ✅
-
- All existing upload files intact ✅
Root cause / timeline for others who hit this:
- Volume config updated to 5000 MB via Railway canvas → Deploy Changes clicked
- Deployment status: SUCCESS immediately
- Host zvol (/sys/block/zdXXX/size) still showed 983040 sectors (~480 MiB) right after deploy
- ~few minutes later, Live Resize propagated automatically — disk expanded to ~4.4 GB
Takeaway: If you update a Railway volume size and the deploy succeeds but df -h still shows the old size, wait a few minutes before panicking — the Live Resize can have a short propagation delay after the deployment completes.
Status changed to Solved hichamjikh8-gif • about 1 month ago