Backblaze support transcripts

The following are the full transcripts of all my support communication with Backblaze technical/customer support. This post is intended as reference for my post Backblaze one year review: 70% data loss unexplained, mediocre software, distrust. All redacted information is identifiable by gray curly brackets, {}.

Backblaze support

While dealing with support teams, I do my best to be respectful, uplifting, and positive in an effort to offset the majority of negative requests they receive. That said, in my dealings with Backblaze, they were usually quick to respond but frequently provided inaccurate or incomplete answers. Not including the final support request involving my lost files, I would rate the support just average.

All dates/times listed are in the America/Toronto timezone.

Index

Request #	Title	Date begins
329150	Failed restore	January 23, 2018
321691	Some files reuploading after recent event/beta client bug?	December 29, 2017
320850	Backup restarted from beginning, now doing dedup slowly, any way to speed it up?	December 26, 2017
297101	Folder that starts with three dots not backed up	September 19, 2017
294635	Notification area icon not being restored bug/feature request	September 9, 2017
272684	Missing files	May 18, 2017
272393	Olark {Backblaze website} chat with {my e-mail address}	May 17, 2017
266722	Frequently stuck producing file lists on pause/resume	April 17, 2017

Request #329150:
Failed restore

January 23, 2018 00:59

Hi there.

I had another drive fail last night, and I’m trying to download my files from Backblaze.

I selected a major folder to restore, and it says:
Selected:
71,838.04 MB

But the restore is only 1.8GB, and in the zip archive most of the files are missing. I stepped back and selected some of the files that are missing manually, and the restore took only a few seconds to prepare and show up as “Failed”.

Is there anything I can do?

January 23, 2018 12:56

Hello,

Thanks for reaching out.

In order to see what’s going on with your backup we’ll need to get the password for your account. Feel free to change this to something you feel comfortable sharing by logging into your account at Backblaze.com, clicking the My Settings link in the left-hand column, and changing your password there.

Best,

{Support 1}

Get to know me! https://www.backblaze.com/blog/{Support 1 link}
The Backblaze Team

January 23, 2018 14:03

Thanks for responding so quickly {Support 1}.

I’ve changed my password for you, so here are my credentials now:
username: {Username}
password: {Password}
I’ve also turned off 2FA.

I’ve done more attempts at restores, and for the most part, no restores seem to be complete. In one case I selected a whole group of files at 75GB, and got back a 128KB zip with a bunch of text files. Trying to restore individual files from those folders results in a ‘Failed’.

As an example, if you attempt to restore T:\Lorax, which is an old video project of mine I did for one of my sister’s plays. The selected files are 8,344.64 MB, but the resulting zip is only 1.55GB, and attempting to restore the main file (also missing from the zip), T:\Lorax\Lorax Youtube 1080p.mp4, results in a Failed response.

Another example, T:\TapesEdit, which are my childhood cassettes I digitized and edited at one point. Selected: 7,409.11 MB, but the resulting zip is 463.7 MB. Attempting to restore one of the missing files in the zip like T:\TapesEdit\Flubber.wav, results in a Failed.

Also of note if it’s important, T: is the drive that’s failed on my side.

Let me know if there’s anything else you need. Thanks in advance for your help 🙂

January 23, 2018 19:33

Hello,

Thanks for your patience while I worked with our devs to look into this.

Please feel free to reset your password back and re-enable 2FV as we will no longer need access to your account.

After further review, it appears that several large files in your backup from the T drive are missing parts from their file. Typically when this happens Backblaze will go through a process called self-healing where the missing parts of a file will be reuploaded. However in your case it looks like 2 files from the 2 drive were never able to be backed up at all for some reason which kept the backup in an initial backup state. The self-healing only occurs once the initial backup is complete, so because your backup never left the initial backup state the self-healing never occurred to fix these files in your Backblaze backup.

Unfortunately because of this we do not have the several files from your T drive in your backup and there is not a way for us recover them as good versions of the files were never uploaded to our servers successfully. What appears when you download folders from the T drive now is the entirety of what is on our servers for the drive. This was a failure of the Backblaze system to work correctly with your computer and drives and if you did want to cancel your Backblaze service we would be happy to offer a full refund of your subscription.

If you would like to continue with us then a fresh backup would be necessary to make sure that files are all backed up correctly. If you need help with this I’m happy to walk you through this process to make sure there are no issues with your backup going forward.

Best Regards,

{Support 1}

Get to know me! https://www.backblaze.com/blog/{Support 1 link}
The Backblaze Team

January 24, 2018 01:32

Hi {Support 1}

Thanks to you and the rest of the team for investigating so far.

I did some further tests tonight, and the data loss is not limited to the T drive. For example, trying to recover a folder of 13.4GB from the D drive, results in a 43.08MB zip file.

In addition, I did manage to determine what’s left of T on Backblaze and I’ve found that only 423GB is still on the server, which means 1,103.12 GB has been lost from T (1 GB = 1024MB). I’ll go through the other drives later on to determine the exact number, depending on what happens next.

That said, I do have some follow-up questions, if you don’t mind.

To begin, you have said that this self-healing didn’t happen because of two stuck files. The thing is, I saw this self-healing process last month – see tickets #320850 and #321691 with {Support 3}. In the first ticket, it restarted the backup from zero and went through all the files. In the second ticket, {they said} that the self-healing and re-uploading I’m seeing is normal, even though the files were being re-uploaded that were already on the server (and recoverable with the same hash), however after that ticket was closed, files that were later being re-uploaded were not able to be recovered (and also had the same Failed state that I’m seeing now), but after they were re-uploaded and processed, those files were once again recoverable. Is that not the self-healing?

Back to the two stuck files. When I saw this in the UI, I tried to determine what those two files were and didn’t have much luck. Looking again through bztransmit23.log, I see “ATTEMPTING TO DECLARE VICTORY on initial upload” and then “BackupSummaryUserWouldSee=Selected_1,508,660_files_/_6,028,700_MB__Remaining_508_files_/_3,728_MB
NOT DECLARING VICTORY on initial upload – too much left”
However, the remaining files that are noted in the log never goes lower than 482, but the UI only shows two (and you also mentioned that there’s two). Looking further, the files are either noted as no longer existing, BZ_FILE_TEMPORARY_FILE_BUSY, or BZ_FILE_TEMPORARY_OTHER. There is also far more than two in every ‘diagnosed’ file, so I have no idea what the two files are. I have zipped/attached my full logs to this message and hopefully you can figure it out better than I.

My second question – is there any explanation for how this happened and why it won’t happen again? From my perspective, this plane fell out of the sky, half the people died, and the solution is to build another one and hope for the best? I already know that at least a terabyte is gone and I’ll gladly provide any info or data you need to better analyze and figure this out, but from this side, it seems like you guys don’t care to know or you do know and it’s some sort of secret.

However, I do have some good news for you. I have another backup of the T drive here, which was done the night before this all happened and I’ve restored it to a new disk tonight. With my data back, I would like to test whether this self-healing solution will work. Is there any way we can do this, or is there really no hope in doing so?

My apologies for the long message, I did my best to edit it down. Thanks in advance for your responses!

-Jon
Attached: {bzlogs-jonathankay.7z} (8 MB)

January 27, 2018 00:22

Hi Jon,

My apologies for the delayed response here.

Your description of self-healing is correct – large files from your backup on our servers are missing chunks of the file and Backblaze goes through the self-healing process to correct this and make files available once more via the Restore browser by attempting to reupload the missing chunks of the file(s) in question.

We can’t say exactly why this happened but it’s likely that either the drive had been malfunctioning in some capacity for some time that could have caused this. The self-healing is attempting to complete but has not been able to and will not be able to because of the drive failure.

I’m glad to hear you were able to recover the data off the T drive. The only way to test this would be to start a new backup which will rebuild the data files that keep track of the hashes of these chunks to prevent this from happening in the future.

Best Regards,

{Support 1}

Get to know me! https://www.backblaze.com/blog/{Support 1 link}
The Backblaze Team

This request is closed for comments. You can create a follow-up.

Status
Solved
Priority
Urgent

Request #321691:
Some files reuploading after recent event/beta client bug?

December 29, 2017 15:35

Hello my Backblaze friends. Earlier in the week I opened a case, ticket #320850. Quick summary of that: I had a process with a memory leak which somehow managed to get Backblaze to start again from the beginning. Although it was recovering, it was doing so very slowly, and {Support 3} instructed me to download the beta version of the client and I was back in business in under 24 hours.

However, just yesterday I had reason to pause the client briefly and afterwards it decided it needed to re-upload a large segment of files (maybe a good 100GB or so). This is not the end of the world but it will take a good while on my connection and is a bit of an annoyance. I restored a few of the files and checked the MD5/SHA-1 against my copies, and they appear to be identical, so I’m not sure why they need to be re-uploaded.

I’m reporting this problem in case there’s something you want me to try or if there’s a bug in this beta client that needs addressing. The compressed log files are well beyond the support forms 20MB limit now after my previous problem, so I’ll need to be more selective or upload elsewhere if you require the logs. However, I used your explainfile utility on one of the files and it showed a few interesting things, such as:
– line 8867 – 20171229044543 – ERROR – bz_done_ INCONSISTENCY_FOUND – 20171229044543 – DetectedDifferentFileId_InstructionMinus

I’ve attached the explainfile results.

Thanks!
Attached: {Backblaze Explainfile.txt} (20 KB)

December 29, 2017 21:15

Hi,

It appears that Backblaze has been performing some “self healing” which is when Backblaze checks what’s in the logs of what’s been transmitted in the past vs. what’s on disk vs. what’s on our servers. When inconsistencies are detected, it schedules the files for upload and then scrubs the logs of the inconsistent transfer.

Normally this process only picks up a few hundred out of the few hundred thousand files a typical backup has. However the amount of data experiencing this implies that a sizable number of files were detected as needing healing. Something about the initial backup of those files wasn’t consistent, so Backblaze is working through them one by one to correct them.

So there’s a few choices here:
You could allow the self healing to keep working away at these files and it may take up to 90 days to complete it in some circumstances
If these files are unimportant to be backed up, you could Exclude them from Backup, and Backblaze should ignore them after about a week or so
You could start a fresh backup, which will likely resolve the issue, but will involve reuploading all files

If you have any questions, please reply and let me know.

Cheers,

{Support 3}
Support Technician
The Backblaze Team

December 31, 2017 14:15

Thanks for responding again {Support 3}. So this is normal behavior? I admit I am a bit curious from a technical perspective what those inconsistences would be, since the hashes are the same of the files, although I am aware that files are more of an abstraction in Backblaze and files are ultimately stored as shards. I assume encryption makes it difficult to reasonably repair or resolve this other than reuploading? Although I am pleased to find the client airs on the side of caution, as there’s nothing worse than finding out a “working” backup has broken files.

With regard to the options, I don’t think I could practically exclude the files without going crazy, as they’re random on disk other than being a certain size. When I noticed it was starting with some 30MB files, and it’s up to 46MB files now. So in real terms that’s mostly small video clips, some FLAC music, a few podcasts I’ve archived, etc. which I definitely want to be stored correctly anyway.

All that said, I think I will let it do its thing for now and I’ll let you know if anything else changes. Thank you for all your hard work!

Take care 🙂

Jonathan

This request is closed for comments. You can create a follow-up.

Status
Solved
Priority
Urgent

Request #320850:
Backup restarted from beginning, now doing dedup slowly, any way to speed it up?

December 26, 2017 14:34

Hello Backblaze people! Happy holidays to you! Although if you’re reading this the holiday is most likely over, but have some chocolate on me 🙂

I arrived home late last night (December 25th) to find that an unrelated third-party process was experiencing a memory leak and was consuming most of the RAM+pagefile. So, I just rebooted the system normally.

After the reboot, I took a look at the Backblaze Control Panel and it indicated that nothing on the PC had been backed up and was starting again from the beginning! Knowing the basics of how Backblaze works, as I expected, it did start going through, scanning files and deduplicating, and that’s what is still doing now. However, this process seems pretty slow and appears to be using very little disk activity. I’ve seen this behaviour before when I lost a drive and it had to recover from that, and it took quite a while. The bztransmit log repeatably says “BzFile::ReadLargeFileIntoBuf_Slowly – called by LoadHashTableWithBzChunkFileIds, pausing for 4 seconds to lighten disk I/O”. Is this what is causing the slowdown?

Is there any trick to getting the Backblaze client to turn off the throttling, or other ways to make this process quicker? I did try increasing the number of threads with no improvement. I’ve attached a 7zip of all my logs in case it’s useful.
Attached: {bzlogs.7z} (20 MB)

December 26, 2017 23:34

Hi,

Thanks for reaching out.

I believe you may be correct in your assumption of what is causing the slowdown. This issue is actually addressed in the current beta release for Backblaze.

I would like to have you first open the Backblaze settings, navigate to the schedule tab, and then change the backup schedule to “Only When I Click”. Once you have changed the setting, close Backblaze, and reboot your system.

After the reboot, download and run the beta installer, available here: http://files.backblaze.com/

No need to uninstall the existing version of Backblaze, simply run the new installer in place. Once the installation is complete, reboot the system once more.

After this second reboot, open the Backblaze settings and change the backup schedule to “Continuous”. Once you have done so, please hold the Alt or Option key and click the “Restore Options…” button in the Backblaze Panel to force an immediate rescan on your machine. Let Backblaze run uninterrupted for at least 24 hours.

Please note that both restarts are vital steps in this process. Please do not skip the reboots in these steps.

Let me know if this doesn’t alleviate the problem.

Cheers,

{Support 3}
Support Technician
The Backblaze Team

December 27, 2017 13:08

Thank you {Support 3}, I have followed your instructions and it is performing significantly faster now. Thank you for all your help!

Jonathan

This request is closed for comments. You can create a follow-up.

Status
Solved
Priority
—

Request #297101:
Folder that starts with three dots not backed up

September 19, 2017 11:31

Hey backblaze people!

I noticed a few days ago that one of my folders wasn’t on Backblaze, when it technically should’ve been uploaded a long time ago. The folder in question contains three ~7MB MP3 files from an album called “…To The Music” and the folder (truncated this a bit for readability) is D:\Music\…To The Music\.

My current guess is that this is being ignored as I believe the uploader ignores folders that start with dots (like your .bzvol folders). Is that correct? I did make an effort to find documentation on whether what I’m seeing is intended or not, but couldn’t find anything definitive.

Also, is there any way to fix this, other than renaming the folder? I took a look at the exclusions xml but didn’t see anything in there that would cover dotted folders.

Thanks in advance for any advice or suggestions! 🙂

September 19, 2017 16:23

Hi;

Thank you for reaching out to us today.

Backblaze has a diagnostic tool called “Explainfile” which can help determine why these files aren’t backed up accurately.

I’ve attached a script to run the diagnostic. Unzip the script to a convenient location, and drag and drop an example file that’s not backed up correctly on to the script. It will output a Backblaze Explainfile.txt to your desktop if it was run correctly.
Note: you can not double click the script to run it, you must drag and drop a file to it.

Please submit the outputted Backblaze Explainfile.txt to your response.

Regards,

{Support 2}
Find out more about me: https://www.backblaze.com/blog/{Support 2 link}
The Backblaze Team
Attached: {Backblaze Explainfile Win.zip} (478 Bytes)

September 19, 2017 22:17

Thanks for all your help {Support 2}. I’ve done that and the results of each file were PRIMARY_DIAGNOSIS: healthy_and_found_in_bz_done.

However, I went back into the View/Restore Files and noticed this magical option, ‘show hidden files’, and sure enough, there was the folder and the MP3 files.

My sincere apologies as I had actually forgotten about this important detail — this folder was in fact set as hidden when I noticed this now-not-so-missing folder on Friday and I “unhid” it at that point. I had checked the bztransmit logs to see if it transferred after removing the hidden attribute as I assumed it was never uploaded, but obviously it had no reason to upload since it was already there.

My follow-up question for you is since the files and folder are there, will my removal of the hidden attribute change on the folder eventually trickle down into my backup or is that not something the uploader covers?

I’ve still attached the file in case it’s helpful 😊

Also, as I write this, {relatable personal information about support 2 from their Backblaze blog entry}, so you’re in good company!

Thanks so much!

Jon

September 20, 2017 16:14

Hi Jon;

I am glad to hear the files were backed up!

If you have changed the folder from being hidden to visible, that should update in the next couple of backups. It may take 48 hours for the change to be reflected in the backup.

I would recommend checking the backup in about 2 days and see if checking the box to not show hidden files/folders makes that folder disappear again.

If the folder is still hidden, let me know.

Best Regards,

{Support 2}
Find out more about me: https://www.backblaze.com/blog/{Support 2 link}
The Backblaze Team

This request is closed for comments. You can create a follow-up.

Status
Solved
Priority
Urgent

As of March 2018, this folder is still shown as hidden.

Request #294635:
Notification area icon not being restored bug/feature request

September 09, 2017 00:46

Greetings Backblaze!

An issue that has been bugging me for quite some time is when Windows Explorer (explorer.exe, ie. the taskbar) crashes or is otherwise restarted in some way, the Backblaze Control Panel icon does restore itself into the notification area/system tray. I admit I was slightly disappointed it wasn’t resolved with the latest release.

I understand that the team is concentrated on the more functional aspects of Backblaze, but handling this simple situation really is expected for any application that runs in the notification area, and the required API/window message to make this happen has existed since IE4/Windows 98, which is why I label this first as a bug.

Adding this function is not complicated and would take just a few minutes, and is well documented in MSDN: https://msdn.microsoft.com/en-us/library/windows/desktop/cc144179.aspx#Taskbar_Creation_Not

One just has to register for the window message and then run your Shell_NotifyIcon initialization again when the message comes.

In terms of repro steps:
1. Have Backblaze running
2. Open up Task Manager
3. Click on Windows Explorer
4. Choose Restart button
5. Note how every other application in the notification area comes back except Backblaze

Alternatively, you can kill the explorer.exe process in Task Manager, and then start it up again, or on Win10, you can hold down Ctrl-Shift, right-click on an empty area of the taskbar and choose Exit Explorer.

Keep up the good work. Thanks so much!

September 09, 2017 14:41

Hello,
Thanks so much for the feedback & detailed info.

I will pass this info along as desired.

1. Does the icon re-appear after about an hour or so?
2. How do you restore the icon after reproducing the behavior?
3. Does restarting also restore the icon?

Please let us know, so we may investigate further.
Thanks,

–{Support 4}
Tech. Support Tech.
The Backblaze Team

September 10, 2017 16:35

Update:
Our dev took a look at it, and was able to squish the bug. He said to tell you ” thanks and it should appear in a build soon. ”

No need to supply the additional info, thanks again.
Sincerely,

–{Support 4}
Tech. Support Tech.
The Backblaze Team

September 11, 2017 10:18

Wow, thanks a lot {Support 4} and everyone else involved. I am very much impressed. I look forward to that in later releases and I think others will benefit from having that sorted out also.

Now about my other feature requests… just kidding 😉
Take care!

This request is closed for comments. You can create a follow-up.

Status
Solved
Priority
—

Request #272684:
Missing files

May 18, 2017 23:43

Hi there. I scanned some pictures earlier in the week, placed them in their own folder and went about my business. According to the C:\ProgramData\Backblaze\bzdata\bzlogs\bzreports_lastfilestransmitted\16.log file, the files were transmitted to the server in the evening on the 16th but although other files (in this case, other scanned images in other folders I did later on) in the log file show up in my account, the one particular folder and its jpg files do not appear to restore. The files also don’t appear if you search.

It’s my understanding that the files should show up within 24 hours.

The missing folder is “D:\Users\TReKiE\Pictures\Scans\High School\DSC 4AI CompSci\An Intro to C” and the files inside should be book 001.jpg, book 002.jpg, etc.

May 19, 2017 15:55

Hello,

Thanks for reaching out!

It’s possible that these files may not have uploaded yet as large files (over 30 MB) only upload once every 48 hours to conserve bandwidth. Please wait at least 48 hours from when the files were added to your computer and check again on your account online.

Best Regards,

{Support 1}

Get to know me! https://www.backblaze.com/blog/{Support 1 link}
The Backblaze Team

May 20, 2017 15:03

Thank you {Support 1}, always glad to hear from you!

I wanted to let you know that I went and checked again this morning at around 09:45 PDT and sure enough, all the missing files are now there. I did check last night too, according to browser history, 22:23 and the files weren’t showing up at that time.

The original batch of files was transmitted (per \bzlogs\bzreports_lastfilestransmitted\16.log) on the 16th at ~17:00 PDT, so some wolfram math makes that about 3.5 days or about 84 hours. The files themselves are just JPGs at 600KB to 1.5MB, but there was ~200 of them. There were some larger files uploaded at the same time and since that were >300MB and those all showed up fine within hours, it was just these images that were missing.

I only tell you this stuff in such detail because I used to have your job, metaphorically speaking, about 10 years ago (just at Streamload/MediaMax, as far as I know, the first “unlimited” storage provider) and so I know how difficult it is to get reliable user feedback on how the system works and how important that information is doing good CS/tech support.

Have a great day!

Jonathan

May 21, 2017 01:36

Hey Jonathan,

Thanks so much for the detailed feedback!

I’ll be passing this along to our devs to make sure we address any bugs present here.

All the best,

{Support 1}

Get to know me! https://www.backblaze.com/blog/{Support 1 link}
The Backblaze Team

This request is closed for comments. You can create a follow-up.

Status
Solved
Priority
Urgent

Request #272393:
Olark {Backblaze website} chat with {my e-mail address}

May 17, 2017 17:30

Location: Canada (Waterloo)
Referred from: https://www.google.ca/
On page: https://www.backblaze.com/help.html
IP address: 99.243.12.29
Talked with: {Support 1}

{Support 1}
03:26 PM | Hi Jonathan, how can I help you today?

Jonathan ({my e-mail address})
03:27 PM | Hi {Support 1}! just a quick question i think, i was watching the uploader yesterday upload some files and they don’t yet appear to restore yet, i do believe that’s normal based on my observations previous but i’m curious how long it should be, just generally, for files to show up.
03:28 PM | appear in the restore file listings i mean

{Support 1}
03:29 PM | Generally it will take about 2-4 hours before a completed upload is available for Restore online, but it can take up to 24 hours before they show as available.

Jonathan ({my e-mail address})
03:29 PM | alrighty, thanks for the info, most appreciated 😀

{Support 1}
03:30 PM | No problem!

Jonathan ({my e-mail address})
03:30 PM | Have a great day 🙂

This request is closed for comments. You can create a follow-up.

Status
Solved
Priority
—

Request #266722:
Frequently stuck producing file lists on pause/resume

April 17, 2017 23:52

When I press the Pause Backup button and then choose Backup Now again, 8 out of 10 times, it will get stuck with “Producing file lists…”, CPU usage on the backblaze bztransmit64.exe is virtually 0 and using process monitor, i can see that it is accessing files, but at an extremely slow rate. To fix the problem, I restart the Backblaze Service, hit Backup Now again, and it now works as expected. In fact this happens so often, I always restart the service prior to resuming it, to make sure it works. This is pretty annoying though, is there anything I can do assist in getting this issue fixed?

Note that I am in the Initial Backup phase.

April 18, 2017 17:58

Hi;

Thanks for writing in!

If you are needing to pause the backup for some reason, change the schedule to Only when click. That will stop the backup. Then when you click “Backup Now” it will start backing up again and you should not see the same issues.

When you click on Backup Now, even under continuously after pausing it, it can take some time for the backup to resume as it needs to scan the machine again to make sure it restarts from the correct spot.

Please let me know if you still encounter any issues.

Regards,

{Support 2}
Find out more about me: https://www.backblaze.com/blog/{Support 2 link}
The Backblaze Team

April 21, 2017 14:59

Hi {Support 2}

I tried this over the past few days to see if it would fix the problem. Although it did seem to be initially successful, last night I paused the backup to send an important large email quickly, and then pressed Backup Now when complete. I then locked the computer and let it do its thing. I had a busy morning/afternoon and now that I check, 13 hours later, Backblaze says “Producing File Lists…” and does not appear to be doing anything. The window indicating that Backblaze is scanning the disk had finished, but never went beyond that.

In other words, I still have the same problem despite the schedule change.

Going over to the Task Manager, Details tab, I saw that there are only two backblaze processes running. bzbui.exe and bzserv.exe. The bztransmit64.exe, which I expect to be there during the Producing File Lists phase of backup, was not running.

I pressed on Pause Backup again to see if I could get it to pause to try and restart it. Although the button selected fine (the bzbui isn’t frozen or anything), it did not pause the backup. So I went back to my original workaround, restarting the Backblaze Service, which then makes the Pause Backup button change back to the Backup Now button, and pressed the button. It then scanned the disk, processed the file lists in under 60 seconds and started sending out the backup again.

Upon checking the Event Log and Problem Reports, bztransmit64 never crashed, in fact no Backblaze process has crashed, so certainly that isn’t the cause of the problem.

I have a relatively normal configuration, a few drives, all NTFS, all single partitioned (other than the Windows drive with its EFI and Recovery partitions). No antivirus or weird security software installed. Backblaze set to automatic throttling. Are there any log files or similar we could be looking at?

Thanks for all your help so far.

Jonathan

April 21, 2017 17:15

Hi Jonathan,
{Support 2} is out right now but I will try to assist in {their} stead.

At this time, I’d recommend setting it back to “Continuously” to ensure it runs as often as possible. For even more info on the different Schedule settings and their functionality, we have this article: https://help.backblaze.com/hc/en-us/articles/217666488-Schedule-Settings-Win-
Then, to ensure the program is “healthy” we can try the following update procedure, which addresses a few different potential issues:
“- Please try disabling any network monitoring programs, firewalls and antivirus on your machine, then follow these steps to repair the program:

1. Go to https://secure.backblaze.com/update.htm
2. Download the installer for your computer.
3. Restart the computer. DO NOT skip this step.
4. When the computer is up and running, open the installer and click install now.

– Once you have done so, please try holding the Alt or Option key and click the Restore Options button in the Backblaze Panel to force an immediate rescan on your machine.

Let Backblaze run uninterrupted for at least 24 hours, and let us know if the issue persists.”

Then, also please keep in mind, the program cycles in it’s tasks every 1-3 hours, longer depending on a few factors, such as machine resources and amount of data (number of files AND amount of data).
Additionally, Computer & Drive Sleep settings will still Backblaze so that it wont run during these periods. For other recommendations we have this article: https://help.backblaze.com/hc/en-us/articles/217664608-Best-Practices

Please let us know if there’s any other questions or issues at all. Screenshots appreciated (One of our favorite screenshot tools here: https://www.take-a-screenshot.org/ ).
Thanks,

–{Support 4}
Tech. Support Tech.
The Backblaze Team

April 24, 2017 03:14

Thank you {Support 4}

I’ve followed your instructions below and I’m afraid there’s been no change. I paused the backup earlier to upload a video to YouTube and after the 30 or so minutes it took to upload the video, I pressed Backup Now and it stuck at “Processing File Lists” with only the two Backblaze processes running – bzbui.exe and bzserv.exe. After restarting the Backblaze service in services.msc, I press Backup Now again and it proceeds as per normal.

My machine is set to the High Performance power setting in Windows and my drives are set to the highest performance setting in their Advanced Power Management (where available). There is no RAID and no external disks.

I’m not really sure what you would like me to screenshot, there isn’t a whole lot to see beyond “Producing File Lists…”. Please let me know if I can provide any further information that may help!

Jonathan

April 25, 2017 18:00

Hi Jonathan;

I would like you to try temporarily disabling your anti-virus and firewall programs

This is typically caused by a security program interfering with Backblaze.
These help article may explain how to do this:
https://help.backblaze.com/hc/en-us/articles/217664708-How-do-I-disable-my-firewall-

http://www.windowscentral.com/how-permanently-disable-windows-defender-windows-10

https://support.microsoft.com/en-us/instantanswers/c9955ad9-1239-4cb2-988c-982f851617ed/turn-windows-firewall-on-or-off

Then try pausing and restarting the backup after a few minutes. If you do not have the issue, it means those programs were stopping Backblaze from working. You will just need to add Backblaze to the approved/exceptions lists.

If you still have the issues, then I would like to collect some logs. I would like to gather some logs to into this further.

Log files are located at:
C:\ProgramData\Backblaze\bzdata\bzlogs\

Please click on this link for how to show hidden files: http://windows.microsoft.com/en-us/windows/show-hidden-files.

Please zip the bzlogs and bzdatacenter folders and attach them to your reply. The logs will contain file names and locations, but no actual files.

Regards,

{Support 2}
Find out more about me: https://www.backblaze.com/blog/{Support 2 link}
The Backblaze Team

This request is closed for comments. You can create a follow-up.

Status
Solved
Priority
Urgent

Note: There is no missing messages past this point as I did not respond to this message. I didn’t choose to invest any more time because I already had a workaround of my own (restarting the service), and they were at the point of blaming the Windows Defender and Windows Firewall, which made no sense given the technical details. Almost a year later (March 2018), this specific problem has become more rare, but still can occur.