Sandbox command reliability is back to normal. Voided work was never billed; re-run anything you are missing on a fresh fork.

An elevated share of commands running in rollout sandboxes is being cut short by infrastructure failures on our side. Any result marked void was not billed and was excluded from its batch - it is a gap in your data, not a wrong answer - and retrying it on a fresh fork is safe. We are investigating the cause. Nodes and jobs run on separate machines and are not affected.

All affected nodes have been recovered, and interruption credits were applied automatically.

A capacity provider issue interrupted some running nodes. We are recovering the affected nodes now. If one of your nodes was interrupted, the interrupted time is credited to your account automatically - you do not need to file anything.

Our sandbox check is passing again: environments prepare, forks start, and commands run normally.

Our continuous check of rollout sandboxes - preparing a real environment, forking a sandbox from it, and running a command inside the fork - is failing, and we are investigating what it caught. Creating environments, forking sandboxes, or running commands in them may be delayed or may fail while this is open. Nodes and jobs run on separate machines and are not affected, and sandbox work that fails on our side is not billed: it is safe to retry once this is resolved.

Our sandbox check is passing again: environments prepare, forks start, and commands run normally.

Our continuous check of rollout sandboxes - preparing a real environment, forking a sandbox from it, and running a command inside the fork - is failing, and we are investigating what it caught. Creating environments, forking sandboxes, or running commands in them may be delayed or may fail while this is open. Nodes and jobs run on separate machines and are not affected, and sandbox work that fails on our side is not billed: it is safe to retry once this is resolved.

Our sandbox check is passing again: environments prepare, forks start, and commands run normally.

Our continuous check of rollout sandboxes - preparing a real environment, forking a sandbox from it, and running a command inside the fork - is failing, and we are investigating what it caught. Creating environments, forking sandboxes, or running commands in them may be delayed or may fail while this is open. Nodes and jobs run on separate machines and are not affected, and sandbox work that fails on our side is not billed: it is safe to retry once this is resolved.

Our stopped-node wake check is passing again: stopped nodes wake normally with their files intact.

Our continuous check that a stopped node wakes with its files intact is failing, and we are investigating what it caught. Running nodes are not affected, and stopped nodes' files remain safe in durable storage - but waking a stopped node may be slower than usual or fail while this is open. If a wake fails, nothing is lost: it is safe to try again once this is resolved.

Our sandbox check is passing again: environments prepare, forks start, and commands run normally.

Our continuous check of rollout sandboxes - preparing a real environment, forking a sandbox from it, and running a command inside the fork - is failing, and we are investigating what it caught. Creating environments, forking sandboxes, or running commands in them may be delayed or may fail while this is open. Nodes and jobs run on separate machines and are not affected, and sandbox work that fails on our side is not billed: it is safe to retry once this is resolved.

Job submissions are working normally again. If one of your jobs failed during this incident, it is safe to resubmit it now.

Some jobs are failing during image preparation because of an internal issue on our side - not because of anything in the job itself. We are investigating. Jobs that are already running are not affected. If a job of yours failed with an image-related error, it is safe to resubmit it once this is resolved.

Job interruptions are back to normal levels.

Some running jobs are being interrupted and returned to the queue more often than we consider acceptable. Interrupted jobs retry automatically - you do not need to resubmit - but affected jobs will take longer to finish than usual.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

Sandbox command reliability is back to normal. Voided work was never billed; re-run anything you are missing on a fresh fork.

An elevated share of commands running in rollout sandboxes is being cut short by infrastructure failures on our side. Any result marked void was not billed and was excluded from its batch - it is a gap in your data, not a wrong answer - and retrying it on a fresh fork is safe. We are investigating the cause. Nodes and jobs run on separate machines and are not affected.

Sandbox command reliability is back to normal. Voided work was never billed; re-run anything you are missing on a fresh fork.

An elevated share of commands running in rollout sandboxes is being cut short by infrastructure failures on our side. Any result marked void was not billed and was excluded from its batch - it is a gap in your data, not a wrong answer - and retrying it on a fresh fork is safe. We are investigating the cause. Nodes and jobs run on separate machines and are not affected.

Sandbox command reliability is back to normal. Voided work was never billed; re-run anything you are missing on a fresh fork.

An elevated share of commands running in rollout sandboxes is being cut short by infrastructure failures on our side. Any result marked void was not billed and was excluded from its batch - it is a gap in your data, not a wrong answer - and retrying it on a fresh fork is safe. We are investigating the cause. Nodes and jobs run on separate machines and are not affected.

Sandbox command reliability is back to normal. Voided work was never billed; re-run anything you are missing on a fresh fork.

An elevated share of commands running in rollout sandboxes is being cut short by infrastructure failures on our side. Any result marked void was not billed and was excluded from its batch - it is a gap in your data, not a wrong answer - and retrying it on a fresh fork is safe. We are investigating the cause. Nodes and jobs run on separate machines and are not affected.

Sandbox command reliability is back to normal. Voided work was never billed; re-run anything you are missing on a fresh fork.

An elevated share of commands running in rollout sandboxes is being cut short by infrastructure failures on our side. Any result marked void was not billed and was excluded from its batch - it is a gap in your data, not a wrong answer - and retrying it on a fresh fork is safe. We are investigating the cause. Nodes and jobs run on separate machines and are not affected.

Sandbox command reliability is back to normal. Voided work was never billed; re-run anything you are missing on a fresh fork.

An elevated share of commands running in rollout sandboxes is being cut short by infrastructure failures on our side. Any result marked void was not billed and was excluded from its batch - it is a gap in your data, not a wrong answer - and retrying it on a fresh fork is safe. We are investigating the cause. Nodes and jobs run on separate machines and are not affected.

Job interruptions are back to normal levels.

Some running jobs are being interrupted and returned to the queue more often than we consider acceptable. Interrupted jobs retry automatically - you do not need to resubmit - but affected jobs will take longer to finish than usual.

Sandbox command reliability is back to normal. Voided work was never billed; re-run anything you are missing on a fresh fork.

An elevated share of commands running in rollout sandboxes is being cut short by infrastructure failures on our side. Any result marked void was not billed and was excluded from its batch - it is a gap in your data, not a wrong answer - and retrying it on a fresh fork is safe. We are investigating the cause. Nodes and jobs run on separate machines and are not affected.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check - a small real job submitted through the same pipeline as customer jobs - is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Our stopped-node wake check is passing again: stopped nodes wake normally with their files intact.

Our continuous check that a stopped node wakes with its files intact is failing, and we are investigating what it caught. Running nodes are not affected, and stopped nodes' files remain safe in durable storage - but waking a stopped node may be slower than usual or fail while this is open. If a wake fails, nothing is lost: it is safe to try again once this is resolved.

Job interruptions are back to normal levels.

Some running jobs are being interrupted and returned to the queue more often than we consider acceptable. Interrupted jobs retry automatically - you do not need to resubmit - but affected jobs will take longer to finish than usual.

Our stopped-node wake check is passing again: stopped nodes wake normally with their files intact.

Our continuous check that a stopped node wakes with its files intact is failing, and we are investigating what it caught. Running nodes are not affected, and stopped nodes' files remain safe in durable storage - but waking a stopped node may be slower than usual or fail while this is open. If a wake fails, nothing is lost: it is safe to try again once this is resolved.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

Usage and billing data has caught up: your usage page is current again, and invoices, receipts, and credits are landing normally. If something you expected from this window is still missing, file a ticket and we will reconcile it.

Usage and billing data is updating more slowly than usual. Recent activity may not appear on your usage page yet, and a new invoice, receipt, or credit may take longer than usual to show up. We are working on it now. Your nodes and jobs are not affected and keep running normally, and nothing is lost: the underlying usage, payments, and credits are all on record, and your pages fill in as we clear the backlog.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Usage and billing data has caught up: your usage page is current again, and invoices, receipts, and credits are landing normally. If something you expected from this window is still missing, file a ticket and we will reconcile it.

Usage and billing data is updating more slowly than usual. Recent activity may not appear on your usage page yet, and a new invoice, receipt, or credit may take longer than usual to show up. We are working on it now. Your nodes and jobs are not affected and keep running normally, and nothing is lost: the underlying usage, payments, and credits are all on record, and your pages fill in as we clear the backlog.

Job interruptions are back to normal levels.

Some running jobs are being interrupted and returned to the queue more often than we consider acceptable. Interrupted jobs retry automatically - you do not need to resubmit - but affected jobs will take longer to finish than usual.

Job interruptions are back to normal levels.

Some running jobs are being interrupted and returned to the queue more often than we consider acceptable. Interrupted jobs retry automatically - you do not need to resubmit - but affected jobs will take longer to finish than usual.

Job submissions are working normally again. If one of your jobs failed during this incident, it is safe to resubmit it now.

Some jobs are failing during image preparation because of an internal issue on our side - not because of anything in the job itself. We are investigating. Jobs that are already running are not affected. If a job of yours failed with an image-related error, it is safe to resubmit it once this is resolved.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check - a small real job submitted through the same pipeline as customer jobs - is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check - a small real job submitted through the same pipeline as customer jobs - is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Our stopped-node wake check is passing again: stopped nodes wake normally with their files intact.

Our continuous check that a stopped node wakes with its files intact is failing, and we are investigating what it caught. Running nodes are not affected, and stopped nodes' files remain safe in durable storage - but waking a stopped node may be slower than usual or fail while this is open. If a wake fails, nothing is lost: it is safe to try again once this is resolved.

Usage and billing data has caught up: your usage page is current again, and invoices, receipts, and credits are landing normally. If something you expected from this window is still missing, file a ticket and we will reconcile it.

Usage and billing data is updating more slowly than usual. Recent activity may not appear on your usage page yet, and a new invoice, receipt, or credit may take longer than usual to show up. We are working on it now. Your nodes and jobs are not affected and keep running normally, and nothing is lost: the underlying usage, payments, and credits are all on record, and your pages fill in as we clear the backlog.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

Our stopped-node wake check is passing again: stopped nodes wake normally with their files intact.

Our continuous check that a stopped node wakes with its files intact is failing, and we are investigating what it caught. Running nodes are not affected, and stopped nodes' files remain safe in durable storage - but waking a stopped node may be slower than usual or fail while this is open. If a wake fails, nothing is lost: it is safe to try again once this is resolved.

Job interruptions are back to normal levels.

Some running jobs are being interrupted and returned to the queue more often than we consider acceptable. Interrupted jobs retry automatically - you do not need to resubmit - but affected jobs will take longer to finish than usual.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Our large-upload and long-build check is passing again: multi-part uploads finalize and long image builds complete normally.

Our continuous check of the large-input path - uploading a multi-part input and building an image whose build runs for many minutes - is failing, and we are investigating what it caught. Small jobs and already-running workloads are not affected. While this is open, uploading a large input may fail partway or need to be retried, and a long image build may fail near the end. Nothing is lost either way: uploaded parts stay put and can be finalized again, and a job whose build fails is not charged for compute it never received.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check - a small real job submitted through the same pipeline as customer jobs - is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

All affected nodes have been recovered, and interruption credits were applied automatically.

A capacity provider issue interrupted some running nodes. We are recovering the affected nodes now. If one of your nodes was interrupted, the interrupted time is credited to your account automatically - you do not need to file anything.

Our stopped-node wake check is passing again: stopped nodes wake normally with their files intact.

Our continuous check that a stopped node wakes with its files intact is failing, and we are investigating what it caught. Running nodes are not affected, and stopped nodes' files remain safe in durable storage - but waking a stopped node may be slower than usual or fail while this is open. If a wake fails, nothing is lost: it is safe to try again once this is resolved.

Job interruptions are back to normal levels.

Some running jobs are being interrupted and returned to the queue more often than we consider acceptable. Interrupted jobs retry automatically - you do not need to resubmit - but affected jobs will take longer to finish than usual.

Usage and billing data has caught up: your usage page is current again, and invoices, receipts, and credits are landing normally. If something you expected from this window is still missing, file a ticket and we will reconcile it.

Usage and billing data is updating more slowly than usual, so recent activity may not appear on your usage page yet. Your workloads are not affected and no usage records are lost; the numbers will catch up on their own.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Our stopped-node wake check is passing again: stopped nodes wake normally with their files intact.

Our continuous check that a stopped node wakes with its files intact is failing, and we are investigating what it caught. Running nodes are not affected, and stopped nodes' files remain safe in durable storage - but waking a stopped node may be slower than usual or fail while this is open. If a wake fails, nothing is lost: it is safe to try again once this is resolved.

Job interruptions are back to normal levels.

Some running jobs are being interrupted and returned to the queue more often than we consider acceptable. Interrupted jobs retry automatically - you do not need to resubmit - but affected jobs will take longer to finish than usual.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Usage reporting has caught up and is current again.

Usage and billing data is updating more slowly than usual, so recent activity may not appear on your usage page yet. Your workloads are not affected and no usage records are lost; the numbers will catch up on their own.

Our stopped-node wake check is passing again: stopped nodes wake normally with their files intact.

Our continuous check that a stopped node wakes with its files intact is failing, and we are investigating what it caught. Running nodes are not affected, and stopped nodes' files remain safe in durable storage - but waking a stopped node may be slower than usual or fail while this is open. If a wake fails, nothing is lost: it is safe to try again once this is resolved.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check - a small real job submitted through the same pipeline as customer jobs - is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check - a small real job submitted through the same pipeline as customer jobs - is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Job submissions are working normally again. If one of your jobs failed during this incident, it is safe to resubmit it now.

Some jobs are failing during image preparation because of an internal issue on our side - not because of anything in the job itself. We are investigating. Jobs that are already running are not affected. If a job of yours failed with an image-related error, it is safe to resubmit it once this is resolved.

Job interruptions are back to normal levels.

Some running jobs are being interrupted and returned to the queue more often than we consider acceptable. Interrupted jobs retry automatically - you do not need to resubmit - but affected jobs will take longer to finish than usual.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check - a small real job submitted through the same pipeline as customer jobs - is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Job submissions are working normally again. If one of your jobs failed during this incident, it is safe to resubmit it now.

Some jobs are failing during image preparation because of an internal issue on our side - not because of anything in the job itself. We are investigating. Jobs that are already running are not affected. If a job of yours failed with an image-related error, it is safe to resubmit it once this is resolved.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Our continuous check that a stopped node wakes with its files intact is failing, and we are investigating what it caught. Running nodes are not affected, and stopped nodes' files remain safe in durable storage - but waking a stopped node may be slower than usual or fail while this is open. If a wake fails, nothing is lost: it is safe to try again once this is resolved.

Our stopped-node wake check is passing again: stopped nodes wake normally with their files intact.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some pages on givemeanode.com are failing to load or are showing an error page. We are investigating. Your nodes and jobs are not affected: everything that is running keeps running, and the API, CLI, and MCP interfaces remain available.

The website is loading normally again. Nodes, jobs, and the API were not affected.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some jobs are failing during image preparation because of an internal issue on our side - not because of anything in the job itself. We are investigating. Jobs that are already running are not affected. If a job of yours failed with an image-related error, it is safe to resubmit it once this is resolved.

Job submissions are working normally again. If one of your jobs failed during this incident, it is safe to resubmit it now.

Our continuous end-to-end check - a small real job submitted through the same pipeline as customer jobs - is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Usage and billing data is updating more slowly than usual, so recent activity may not appear on your usage page yet. Your workloads are not affected and no usage records are lost; the numbers will catch up on their own.

Usage reporting has caught up and is current again.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check — a small real job submitted through the same pipeline as customer jobs — is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check — a small real job submitted through the same pipeline as customer jobs — is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Our continuous end-to-end check — a small real job submitted through the same pipeline as customer jobs — is failing, and we are investigating what it caught. Jobs you submit may be delayed or may fail while this is open. Queued jobs keep their place in line, and jobs that cannot run are not charged for compute they never received.

Our end-to-end check is passing again: jobs are building and running normally through the whole pipeline.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Node provisioning is back to normal. New nodes are becoming ready at the usual speed.

Some new nodes are taking longer than usual to become ready. We are investigating. Nodes that are already running are not affected, and pending nodes will come up without any action on your part.

All affected nodes have been recovered, and interruption credits were applied automatically.

A capacity provider issue interrupted some running nodes. We are recovering the affected nodes now. If one of your nodes was interrupted, the interrupted time is credited to your account automatically — you do not need to file anything.

Usage and billing data is updating more slowly than usual, so recent activity may not appear on your usage page yet. Your workloads are not affected and no usage records are lost; the numbers will catch up on their own.

Usage reporting has caught up and is current again.