Alibaba Cloud foreign card top up Fixing 'Disk Quota Exceeded' Errors on Alibaba Cloud Linux Instances

If your Alibaba Cloud Linux instance suddenly stops writing files, database logs fail, or deployments break with Disk quota exceeded, the real problem is usually not just "disk is full." In practice, I see three different situations behind this message: the system partition is full, an inode quota has been hit, or a user or application quota is enforced on a mounted volume. The fastest fix depends on which one you are actually hitting.

Alibaba Cloud foreign card top up There is also a less obvious issue: many people cannot complete the repair in time because their cloud account cannot be topped up, the renewal fails, or the account is under risk control review. So this article focuses on both sides of the problem: how to restore the instance and how to avoid getting blocked by account, payment, or compliance issues when you need to act quickly.

What users usually mean when they search this error

In real operations, "Disk Quota Exceeded" normally shows up in one of these cases:

• Web uploads fail and the application returns a 500 error.
• Database or log services cannot write to disk.
• Package installation stops halfway with a quota-related message.
• SSH login works, but creating new files or editing existing files fails.
• The instance still has some free space, but the error appears for one specific user or directory.

The first thing to avoid is guessing. On Alibaba Cloud Linux, the message can be triggered by filesystem usage, inode exhaustion, user quota, or a cloud disk that is smaller than expected after a resize operation that was not completed inside the guest OS.

Fast triage: what to check first

If the service is down and you need a quick answer, use this order. It saves time and prevents unnecessary changes.

1. Check total disk usage with df -h.
2. Check inode usage with df -i.
3. Check whether the error is user-specific by switching to root or another account.
4. Check the mounted data disk and log directories separately.
5. Check for quota configuration with quota -v or mount options if quotas are enabled.

If df -h shows 100% usage on / or a data partition, the fix is usually capacity cleanup or disk expansion. If df -i is 100%, you have too many small files, not too much data volume. If only one user or one application path fails, quota rules are probably in place.

Most common root causes on Alibaba Cloud Linux instances

Alibaba Cloud foreign card top up Step 1: Confirm whether this is a real disk shortage

Start with the basics:

df -h
df -i
mount
lsblk

Here is how I interpret the output in practice:

• If / is full, the OS disk is the issue, and the instance may become unstable quickly.
• If /data or /var is full, application services often fail first.
• Alibaba Cloud foreign card top up If inodes are 100% but space is not, the directory likely contains millions of tiny files.
• If the disk was enlarged in the console but Linux still shows the old size, the guest OS has not been extended yet.

I have seen many teams resize a cloud disk in Alibaba Cloud, assume the issue is fixed, and then keep getting the same error because the partition and filesystem were never expanded. That is a common operational mistake.

Step 2: Clean up safely before you expand

If the instance is production and you need a fast recovery, cleanup is usually the lowest-risk first move. Focus on files that are large, temporary, or easy to regenerate.

• Rotate or compress application logs.
• Remove old deployment packages and build artifacts.
• Clear package manager caches.
• Prune unused Docker images, containers, and volumes if the host runs containers.
• Move backup files and exports to Object Storage Service if they do not need to stay on the instance.

Useful checks:

du -sh /* 2>/dev/null
du -sh /var/* 2>/dev/null
du -sh /data/* 2>/dev/null
journalctl --disk-usage

If the machine hosts a database, be careful with manual cleanup. Removing the wrong file in /var/lib or a database data directory can turn a quota issue into a data loss incident. In that case, clear logs first, not data files.

Step 3: If the error is quota-related, check the actual limit

Some Alibaba Cloud Linux instances are configured with quotas at the filesystem or user level. That is common on shared data volumes, hosting environments, and systems where multiple applications run under different users.

Check:

quota -v
repquota -a

If quotas are enabled, the fix is not just "free some space." You may need to:

• Raise the soft or hard quota for the user or project.
• Move files to a different mount point with more allowance.
• Adjust the application to write to the correct directory.
• Coordinate with the server admin if you are not the owner of the quota policy.

For hosted environments, I often see the web server user hit quota first while root still has space. That creates confusion because the machine looks healthy from the administrator's shell, but the application still cannot write.

Step 4: Expand the cloud disk the right way

Alibaba Cloud foreign card top up If you are already close to capacity or the workload is growing, expansion is the durable fix. On Alibaba Cloud, that usually means increasing the cloud disk size in the console, then extending the partition and filesystem in the Linux guest.

The important point is this: increasing disk size in the console does not automatically update the filesystem inside the instance.

Typical sequence:

1. Back up critical data first.
2. Expand the cloud disk in the Alibaba Cloud console.
3. Use lsblk to confirm the new device size.
4. Extend the partition if the disk uses partitions.
5. Grow the filesystem, such as ext4 or xfs.

If you are not comfortable with partition changes on a live system, schedule a maintenance window. For a busy production node, a rushed partition command can cause more downtime than the original quota issue.

When the real problem is account funding or renewal, not Linux

This is where many teams get stuck. They know the disk must be expanded, but the cloud account cannot place the order, renew the instance, or pay the invoice because funding is blocked. In practice, the repair path can fail for reasons that have nothing to do with the server itself.

What usually blocks the purchase or renewal

• The account has no valid payment method attached.
• The card is issued in a region that the platform flags for additional review.
• The account has not completed identity verification, so spending limits remain restricted.
• The account is under risk control and cannot create new resources or top up immediately.
• The enterprise verification is incomplete, so invoice or quota-related operations are delayed.

From a practical standpoint, if you are trying to solve a disk quota issue on a production instance, you should confirm your account can actually support the fix before the instance reaches a hard stop.

Account purchase and KYC: what matters before you need urgent capacity

A lot of cloud buyers only think about KYC after the account is already under pressure. That is too late. If you operate business systems, the safe approach is to finish identity verification early, bind a valid payment method, and make sure the account can renew without manual intervention.

In Alibaba Cloud International, identity verification and risk screening can affect what you are allowed to buy, how much you can spend, and whether the account can pass a sudden top-up or renewal. The exact checks vary by region, entity type, and payment method, but the practical result is consistent: if verification is incomplete, you may not be able to expand storage when you need it most.

What I recommend for new buyers

• Verify the account before production goes live.
• Use a payment method that supports recurring charges and higher limits.
• Keep a small balance or spending headroom for emergency disk expansion.
• Document who can approve renewals if the account is enterprise-managed.

A common failure pattern is this: the server is already at 95% disk usage, the admin requests a larger disk, and the payment card fails verification or the account is temporarily blocked by risk control. At that point, the operational issue becomes a financial and compliance issue.

Payment methods: why the cheapest-looking option is not always the safest

When comparing payment methods, do not look only at fees. Look at approval speed, charge success rate, and whether the method will still work under time pressure.

For production Linux instances, I usually recommend at least one payment method that can clear immediately and one backup path for enterprise procurement. The cost difference is often smaller than the business cost of delayed disk expansion.

Alibaba Cloud foreign card top up Risk control and compliance reviews: the hidden delay nobody plans for

Alibaba Cloud accounts can enter risk control review for reasons that are not obvious to the user:

• New account with immediate high-value purchases.
• Alibaba Cloud foreign card top up Payment method mismatch with billing profile region.
• Alibaba Cloud foreign card top up Multiple failed payment attempts in a short period.
• Rapid creation and deletion of resources.
• Inconsistent identity or enterprise data.

If this happens while you are trying to fix a disk shortage, the practical response is to stop retrying random card payments. That usually makes the review slower. Instead:

1. Check the account notice or billing dashboard for the exact restriction.
2. Prepare identity or company documents if they are requested.
3. Use the same business name, billing address, and payment profile consistently.
4. Keep communication concise and factual when asking support to review the hold.

In my experience, accounts that are set up cleanly from the start clear these reviews faster than accounts with repeated billing mismatches and multiple failed payment attempts.

Usage restrictions that affect storage fixes

A disk quota issue can be made worse by account-level restrictions. For example:

• You may be able to log in to the console but not create a larger disk.
• The instance may be allowed to run but blocked from additional purchases.
• Renewal reminders may not help if the account is already restricted for compliance reasons.
• Enterprise approval workflows may delay changes even when the technical fix is simple.

This is why teams that run business-critical instances should not wait until the last day of a billing cycle to renew. If your root disk is already tight, leaving renewals to the last hour creates an avoidable outage risk.

Cost comparison: clean up, resize, or move workloads

The cheapest fix is not always the best one. Here is how I usually compare the options:

If you run a database, email service, or log-heavy application, separating the system disk from the data disk often lowers operational risk. When the root partition fills up, the whole host can become hard to manage. Keeping data on a dedicated volume gives you more room to recover.

A practical recovery path for production incidents

If the instance is already failing, I would handle it in this order:

1. Confirm whether the outage is caused by disk space, inode exhaustion, or quota limits.
2. Free the safest space first: logs, cache, and temporary files.
3. Move critical but non-active files to OSS or another disk.
4. Expand the disk if usage is structurally high.
5. Alibaba Cloud foreign card top up Fix the application path so the issue does not come back after the next deployment.
6. Check the cloud account status so renewals and future expansions will not fail.

If the account is blocked and you cannot buy extra capacity, the fallback is to reduce disk consumption immediately and buy time. That can mean log truncation, cache cleanup, or moving archived files off the instance while support or finance clears the payment issue.

Real-world example: the error was not where the team thought

One common case I have seen on Alibaba Cloud Linux is a small application server with a 40 GB system disk. The team saw Disk quota exceeded in the application logs and assumed the database partition was full. In reality, the issue was not the main data volume. The web application user was writing session files to /tmp, and /tmp was on the root partition, which had filled up from log growth and package caches.

The fix was not a database migration. It was:

• Move session files to a dedicated data volume.
• Rotate and compress application logs.
• Increase the root disk by a modest amount.
• Set up a monthly review so the same pattern did not repeat.

This kind of incident is why I always check the exact path and user context before changing the disk layout.

FAQ

Why does the server still show free space but writes fail?

Usually because you are hitting inode limits or a user quota, not total block space. Check df -i and quota settings before expanding the disk.

Does increasing the disk size in the Alibaba Cloud console fix the error immediately?

No. You still need to extend the partition and filesystem inside the Linux instance. If you skip that step, the OS keeps using the old size.

Can I solve this by upgrading the account or adding another payment method?

Not directly, but it matters if you need to buy more disk or renew the instance. If the account is not verified or the payment method fails, the technical fix may be delayed.

What if the account is under risk control review?

Stop retrying random payments. Check the billing or security notices, prepare the required identity or company documents, and contact support with clear account details. Repeated failed attempts often make the review slower.

Is it cheaper to clean up or expand storage?

Cleanup is cheaper today, but expansion is often cheaper operationally if the workload keeps growing. For log-heavy or database workloads, the right answer is usually a combination of cleanup now and a planned resize soon.

How do I avoid this in the future?

Set alerts for disk usage and inode usage, rotate logs, separate system and data disks, and keep the cloud account ready for renewals and emergency purchases. The best prevention is not only technical monitoring, but also making sure billing and verification are already in place before an incident.

What to do next

If you are dealing with this error right now, the best path is to confirm the exact bottleneck first, then decide whether cleanup, quota adjustment, or disk expansion is the safest fix. At the same time, make sure your cloud account can actually complete the order or renewal if expansion is needed.

In practice, the strongest operations teams treat storage problems and account readiness as one workflow. That is what keeps a routine disk issue from turning into a business outage.