New backup system on 11.38 corrupt files

[Gareth-AWD]

I ran a backup manually yesterday and noticed the generated .gz file was half the size of the legacy backups so I downloaded it and according to both my compression software the files are corrupt.

Last nights backup was also a different size again to the backup I ran manually.

Any ideas?

 

[ThinIce]

Does the backup log contain any useful information or does everything there appear okay?

 

[Infopro]

I ran a backup manually yesterday and noticed the generated .gz file was half the size of the legacy backups so I downloaded it and according to both my compression software the files are corrupt.

Last nights backup was also a different size again to the backup I ran manually.

Any ideas?

One. Please open a ticket about this and get cPanel Technical Support in there for a closer look.

Thanks

 

[cPanelKenneth]

I ran a backup manually yesterday and noticed the generated .gz file was half the size of the legacy backups so I downloaded it and according to both my compression software the files are corrupt.

Last nights backup was also a different size again to the backup I ran manually.

Any ideas?

https://support.cpanel.net

As Infopro encouraged, please open a support ticket on this matter. We'd like to get to the bottom of it, and resolve any issues, post haste.

Thank you.

 

[Gareth-AWD]

https://support.cpanel.net

As Infopro encouraged, please open a support ticket on this matter. We'd like to get to the bottom of it, and resolve any issues, post haste.

Thank you.

I will do Monday, I'm away at the moment and I've switched back to legacy backups. I have 5 cpanel servers so am keen to get the new backup system working on all, but this one server is on the current tier where as the others are on RELEASE.

This particular server only has 5 accounts. I'm not sure if its the pigz enabling that does it but will also run a test enabled and disabled with the new system.

One final question, my FTP location doesn't seem to have the config backup only the accounts backup, I thought that was FTP'd.

 

[cPanelKenneth]

I will do Monday, I'm away at the moment and I've switched back to legacy backups. I have 5 cpanel servers so am keen to get the new backup system working on all, but this one server is on the current tier where as the others are on RELEASE.

This particular server only has 5 accounts. I'm not sure if its the pigz enabling that does it but will also run a test enabled and disabled with the new system.

One final question, my FTP location doesn't seem to have the config backup only the accounts backup, I thought that was FTP'd.

Thank you for your interest and input on the new backup system. We await your support ticket so we can get to the source of the matter you are experiencing.

Due to the sensitivity of the files included in the System Backup, an extra step is required to include those when copying to a remote destination. Please see System Backups for information on including these files.

 

[Gareth-AWD]

Ticket now submitted. Out of interest, has the support staff been briefed on the net system or does it go to a higher level because of the nature of it?

 

[Gareth-AWD]

For anyone following this. It appears from the ticket that there is a problem with FTP transfer process rather than the backup file locally.

 

[Gareth-AWD]

Confirmed by cPanel support that this is a bug and has now gone over to the development team and a case created. Outstanding support from all the people who dealt with my ticket. Thank you.

 

[sp3ctre69]

Yeah, nasty bug, especially as it results in a "backup successful" message. Hopefully it'll get sorted soon

 

[sp3ctre69]

Not to seem too alarmist, would it be a good idea for cpanel to send out an email to those on >=Current tiers to not use the new backup system with FTP, as unless you happened to check you may not even realise the backups are not complete? I realise it's unlikely to get fixed in 24 hours, so I guess the advice to those FTP users should really be to go back to the legacy system.

 

[Gareth-AWD]

Not to seem too alarmist, would it be a good idea for cpanel to send out an email to those on >=Current tiers to not use the new backup system with FTP, as unless you happened to check you may not even realise the backups are not complete? I realise it's unlikely to get fixed in 24 hours, so I guess the advice to those FTP users should really be to go back to the legacy system.

Even though CURRENT is isn't probably recommended for some people, I think if it can't be fixed in the next 24 hours from now then a notification does need to go out. Can you imagine if you need to recover a site to discover the only backup you have is incomplete!

 

[sp3ctre69]

Even though CURRENT is isn't probably recommended for some people, I think if it can't be fixed in the next 24 hours from now then a notification does need to go out. Can you imagine if you need to recover a site to discover the only backup you have is incomplete!

It'll be interesting to hear where they are with it today. Apparently there are several developers working on it. I agree though, it makes no difference what tier it's on, but if there is knowledge that something definitely does not work it needs to be communicated.

 

[Gareth-AWD]

It'll be interesting to hear where they are with it today. Apparently there are several developers working on it. I agree though, it makes no difference what tier it's on, but if there is knowledge that something definitely does not work it needs to be communicated.

I know it's being worked on but good that its a number of developers. I guess this is pretty important to get right. Gotta hand it to them for jumping on it.

 

[sp3ctre69]

I know it's being worked on but good that its a number of developers. I guess this is pretty important to get right. Gotta hand it to them for jumping on it.

Has anyone heard anything more from WHM, it's all gone quiet?! I have already had to tell some folks I know who are on 11.38 not to trust the new backup system (they had no idea until they checked their backup sizes and were quite shocked!). Bit shocking that it should be my job to do that! This either needs fixing or an announcement going out, keeping quiet while it is broken isn't professional at all and somewhat damages my trust in WHM.

 

[Gareth-AWD]

Has anyone heard anything more from WHM, it's all gone quiet?! I have already had to tell some folks I know who are on 11.38 not to trust the new backup system (they had no idea until they checked their backup sizes and were quite shocked!). Bit shocking that it should be my job to do that! This either needs fixing or an announcement going out, keeping quiet while it is broken isn't professional at all and somewhat damages my trust in WHM.

I replied to my ticket today but no response asking. I think the ticket is in the hands of the developers so I'm not fretting getting a reply.

The old backup system still works well but see where you're coming from where people need to know not to use it.

I'm also not sure it affects everyone, it might be certain settings or configurations. How big are your backups that are failing as all mine under 2Gb are fine, just my 6Gb site that comes out as 3Gb on FTP. To be honest, I only noticed because I have such an interested in the new system I was comparing pigz speed compared to the old backups.

 

[cPanelKenneth]

I replied to my ticket today but no response asking. I think the ticket is in the hands of the developers so I'm not fretting getting a reply.

The old backup system still works well but see where you're coming from where people need to know not to use it.

I'm also not sure it affects everyone, it might be certain settings or configurations. How big are your backups that are failing as all mine under 2Gb are fine, just my 6Gb site that comes out as 3Gb on FTP. To be honest, I only noticed because I have such an interested in the new system I was comparing pigz speed compared to the old backups.

Hello, Gareth,

Thank you for opening the support ticket with us. The problem appears to stem from a time-out being encountered by the sub-system that handles copying the packages to the remote destination. When the time out occurred, it encountered some unhandled error conditions, which resulted in the corrupt files you saw.

I see the case number was changed to 67985, and the problem should be fixed in 11.38.0.8+.

 

[kisonay]

I am also experiencing this (it typically only happens with larger backup archives) and appreciate the staff looking into it.

do we have an ETA on 11.38.0.8+.

 

[Gareth-AWD]

Hi Kenneth,

Thank you for the update. I was given two case numbers in my ticket 67985 and 68049. Out of interest, what's the difference? Are they both responsible for the corrupt backups?

 

[sp3ctre69]

I am also experiencing this (it typically only happens with larger backup archives) and appreciate the staff looking into it.

do we have an ETA on 11.38.0.8+.

Also interested to know when we can expect 11.38.0.8