A Contabo snapshot is useful as a rollback point, but it is not a complete backup for a 24/7 YouTube stream. Auto Backup is a separate feature available only on eligible VPS families, and independently stored copies of your media and configuration give you another recovery route.
None of these copies keeps your stream live while a VPS restore is under way. Plan for the interruption, rehearse recovery outside an important broadcast, and confirm that the recovered system can actually send a healthy stream to YouTube.
What a 24/7 stream needs protected
A looping channel depends on more than the video file. It needs the source media, the playlist or loop order, encoder and streaming application settings, any scripts or scheduled tasks that start the broadcast, and the VPS operating-system configuration those pieces rely on. If you cannot reproduce one of these after a failure, you may restore the server but still not restore the channel.
Credentials need their own careful treatment. Your YouTube stream key is a secret, not ordinary configuration to place in an unprotected archive or shared document. Record where it is managed and how an authorised operator can retrieve or replace it, without putting the key itself somewhere that exposes it. Keep account recovery information separate and protected as well.
There is also a continuity requirement that backups cannot satisfy: a working route from encoder to YouTube. A copy of your media will not keep the ingest connection open during a server failure, and a restored system may have a changed application state or need a fresh connection. Think of data recovery and stream resumption as two related but distinct jobs.
For a channel based on recorded lessons, for example, keep the original class recordings and a written note of the intended playback order outside the VPS. For a devotional channel, include the approved audio and image files and any schedule that determines what plays next. A practical starting point for the wider workflow is this guide to building a 24/7 YouTube stream for an Indian education channel.
Three different continuity layers
Snapshots, Contabo Auto Backup and independent copies serve different purposes. A snapshot is a convenient saved system state for rolling back after a change. Auto Backup is a scheduled, off-server copy of the VPS with limited retention and whole-VPS restore behaviour. Independent copies are files or records you keep through a separate method, so you are not relying only on the VPS or its provider-held recovery options.
Contabo explicitly says its snapshot feature is not a backup. It saves the current system state for rollback, is managed in the Customer Control Panel, and cannot be downloaded. The provider documents a 30-day expiry and plan-dependent snapshot slots; check the current panel because the deletion date and available slots can vary. This makes a snapshot useful before a risky software change, but a poor sole answer to loss of the VPS or a need to retrieve one file.
Auto Backup has a different role. Contabo documents it as an add-on for Core VPS and Performance VPS, stored in a Contabo data centre separate from the VPS host. It can help if the host or VPS contents are lost, but the copies remain within Contabo’s service and are not a customer-selected independent archive. A restore overwrites the current VPS and returns the whole VPS, not an individual file.
Independent copies are the layer you control separately: original media on a local drive, a second copy in an independently managed storage account, and a protected export of configuration and setup notes. These copies are not automatically current or usable simply because they exist. Update them after meaningful changes and periodically verify that files open and credentials can be accessed by the right person.
| Method | Useful for | Important limit |
|---|---|---|
| Snapshot | Rolling back a system after a change | Not a backup; expires, cannot be downloaded, and slot count depends on plan |
| Auto Backup | Recovering a whole eligible VPS from a recent provider-held copy | Short retention; restore overwrites the VPS, causes downtime and is not a file-level restore |
| Independent copies | Rebuilding or recovering important media and configuration through another route | You must maintain, protect and test the copies yourself |
The sensible choice is usually not one of these instead of the others. Use a snapshot for a near-term change when your plan supports it, consider Auto Backup if its scope and retention suit you, and keep irreplaceable content and setup information independently. See Contabo’s snapshot guidance and VPS Auto Backup documentation for the current provider descriptions.
Check eligibility, retention and limits
Do not assume that every Contabo VPS includes Auto Backup or snapshots. Contabo’s current VPS documentation lists Auto Backup for Core VPS and Performance VPS; Storage VPS is not listed as having snapshots or the Auto Backup add-on. The plan-family comparison also lists one to three snapshot slots for Core VPS plans, five for Performance VPS plans, and none for Storage VPS. Confirm the selected plan’s current features in the provider’s documentation and Control Panel before building a recovery plan around them.
Contabo documents Auto Backup as daily. The first copy is full and later copies are incremental, with at least eight hours between backups. Up to ten copies are retained, each for a maximum of ten days; when the cap is reached, the oldest is removed first. You cannot select the exact time within the provider’s standard backup window or choose the storage location. Those details matter if your channel has a busy period or if you assumed a copy from a particular moment would remain available.
These are provider-documented product limits, not a promise that every recent change will be recoverable. Check the Control Panel to confirm that the feature is enabled and that a backup has completed. Note when the latest copy was made and what work has happened since. A new video or changed playlist created after that copy may not be present in a restore, so maintain those materials separately rather than treating the retention window as an archive.
Storage VPS deserves a separate decision. Contabo positions it for workloads where capacity matters more than peak I/O throughput, including backup and archive use cases, and its documentation distinguishes its storage characteristics from Cloud VPS. That does not make it a required companion server, nor does it give the streaming VPS the same built-in recovery features. If you use a separate storage role, decide how files are copied, protected and restored rather than assuming that another VPS is automatically an independent backup.
Choose what to save: configuration and media
Start with the material that would be difficult or costly to recreate. Keep original video and audio files, still images, subtitles, playlists, and any prepared transitions or overlays. If your stream loops a long programme, preserve the source files as well as the playlist: a rendered or downloaded copy of the final output may not retain the structure you need to revise it later.
Next, document how the stream is assembled. Record the operating-system version, relevant application names and versions, encoder profile, resolution, frame rate, bitrate, audio settings, file paths, and the method used to start or restart the stream. Add instructions for any scheduled job or service, firewall rule, and monitoring check that is necessary. Avoid copying secrets into this document. Instead, indicate where authorised staff can obtain them securely.
Be selective about full-system copies. A whole-VPS backup can save time when the entire machine must be recovered, but it also preserves the system as it was at that moment. It may include outdated software, a broken configuration, or sensitive files that should not be exposed. Independent copies of clean source media and concise setup notes can make a rebuild easier to understand than relying only on an old system image.
Keep the separate copy genuinely separate from the VPS failure you are trying to address. A folder on the same VPS disk is not another recovery route. A local copy helps if the VPS disappears, but may not help if the local computer fails or is unavailable to the person restoring the channel. Choose a method that fits your capacity, access controls and operating practice, then confirm that another authorised operator can reach the necessary files.
You can use a simple inventory: item, current location, second-copy location, last checked, and access owner. Include the stream key only as a reference to its protected storage, never as plain text in the inventory. If you change the encoder or switch the programme order, update the notes and the relevant copies at the same time. This small discipline reduces the chance that recovery instructions describe a setup you no longer run.
If the operational burden is mainly keeping a computer online to play an uploaded file continuously, StreamNeo can remove that specific burden by running the uploaded video as a 24/7 YouTube live stream while your own computer is switched off. It does not replace your responsibility to keep source files and account credentials safe or to test that a stream is behaving as intended.
Understand restore downtime and trade-offs
A VPS restore is a maintenance event, not a seamless handover. Contabo says an Auto Backup restore overwrites the current VPS data and makes the VPS unavailable until the operation completes. Since the restore covers the whole VPS, you cannot use it to recover just a deleted playlist while leaving everything else untouched. Changes made after the chosen backup can be lost, so identify what needs preserving before you start.
For a live channel, unavailability means the encoder cannot keep sending from that VPS during the restore. The stream may stop or lose its connection, and you should expect to reconnect and verify it afterwards. Do not tell viewers that a restore will happen invisibly, and do not schedule one during a broadcast you cannot interrupt. Contabo also says it does not guarantee recoverability or perform recovery for the customer, so having a backup listed in a panel is not proof that recovery will work.
Snapshots carry a different trade-off. They are quick rollback points in the provider’s snapshot workflow, but they expire, have plan-dependent capacity and cannot be downloaded. If the VPS or its associated service is unavailable, the snapshot is not a separate file archive you can carry to another environment. Use it for the job it supports, such as undoing a problematic change, rather than as the entire continuity plan.
Independent file copies take more care to organise and may not capture the complete running system. In exchange, they give you access to the source material and documentation without relying on a whole-VPS restore. A rebuild from those copies can take longer than restoring a recent system state, while a stale whole-system image can be less useful than expected. Match the method to the failure you are preparing for.
If the exact timing of an interruption matters, arrange a maintenance window and communicate it to viewers in plain terms. For a small devotional station, that may mean announcing a brief planned pause and confirming the next programme afterwards. For a local news loop, you may decide a spare playback path or a separate operator is worth the extra complexity. There is no backup setting that removes the need to choose how your channel handles an outage.
Rehearse recovery outside a critical period
A recovery test answers questions that a successful backup job cannot. Can you access the Control Panel? Do you know which restore point to choose? Can you retrieve the independent media copy and configuration notes? Does the restored or rebuilt system have the files and settings it needs? Run the rehearsal outside a high-stakes live period, with enough time to investigate rather than treating it as a quick checkbox.
Because a production restore takes the VPS out of service and overwrites its current data, do not use a live VPS as a casual test target. Where practical, use a separate test environment or rehearse a rebuild from independently held files. If you need to test a production restore, make the interruption an explicit maintenance event and first ensure any newer content or configuration is separately saved. Follow the provider’s current instructions and confirm the specific restore target before committing.
Write down the sequence in the order you would need it during a real incident: who can access the account, where the latest independent media copy is, how to restore or reinstall the playback application, where settings are recorded, and how to reconnect to YouTube securely. The goal is not to script every possible failure but to remove avoidable uncertainty. Give the notes to a second trusted operator and ask them to find the files without your help.
During the rehearsal, compare the recovered system with the inventory. Confirm that the expected programme files exist, the playback order is right, the encoder settings match the intended quality, and the process starts after a reboot if that is part of your design. Record anything that was missing or unclear, correct the copies and instructions, then repeat the relevant part. A test that finds a gap is useful only if you close it.
Verify the stream after recovery
A restored VPS is not a confirmed working channel. First check the operating system and streaming application, confirm that the expected media and configuration are present, and inspect resource and network behaviour while the encoder runs. Then initiate a test broadcast or other controlled check before treating the setup as ready for continuous service.
YouTube recommends testing before a live event and monitoring stream health. Its encoder guidance recommends RTMPS for encrypted ingest and gives H.264 bitrate ranges, including 5–14 Mbps for 1080p at 30 frames per second and 6–17 Mbps at 1080p at 60 frames per second. These are YouTube recommendations for encoder settings, not evidence that a particular VPS or network route can sustain them. Choose a quality target, then test the actual workload on the actual VPS.
Check the live preview or stream health indicators for dropped frames, connection interruptions, audio problems and an incorrect picture. Listen as well as watch: a file can be present yet silent, or the wrong audio track may be playing. A controlled test also gives you a chance to confirm that the intended title, thumbnail and programme are being sent before you return to the regular schedule. YouTube’s live encoder settings and bitrates and stream health guidance are useful references; consult the current official pages because platform guidance can change.
For a channel that relies on OBS, scheduled playlists or a separate playback script, check the whole chain rather than only the encoder process. The health metrics troubleshooting guide can help you distinguish a healthy local process from a healthy outgoing stream. If you need to inspect the signal before it reaches YouTube, the guide to checking an RTMP stream locally covers that separate diagnostic step.
Only after these checks should you return the channel to its normal continuous schedule. Keep an eye on the stream after reconnection, because a successful test at one moment does not establish that the encoder, VPS resources and network will behave indefinitely. If the channel is important to viewers, document who checks it and what they should do if the feed drops again.
Before committing, compare the operating options on the pricing page. When the file and channel are ready, start free — 24-hour trial, no card.
FAQ
Are Contabo snapshots enough to protect a 24/7 YouTube stream?
No. Contabo describes snapshots as saved system states for rollback, not as backups; they expire, cannot be downloaded and have plan-dependent slot limits. Keep independent copies of important media and configuration, and consider Auto Backup only if your plan is eligible and its restore model suits you.
Does every Contabo VPS include Auto Backup?
No. Contabo’s current VPS documentation lists Auto Backup as an add-on for Core VPS and Performance VPS, not Storage VPS. Check the current plan details and Control Panel rather than assuming the feature is enabled or available.
Will my stream stay live while Contabo restores a backup?
No. Contabo says the VPS is unavailable until an Auto Backup restore completes, and the restore overwrites the VPS with a whole-server copy. Plan for an interruption, then reconnect and test the broadcast after recovery.
How do I know the recovered setup is ready for continuous streaming?
Check that the media, playback order, configuration and credentials are available, then run a controlled test at your intended resolution, frame rate and bitrate. Monitor YouTube stream health and the VPS’s resource and network behaviour; a completed restore alone does not prove that the stream can resume.