Disk Report Error Flag -- What does it indicate

Martin, I have not seen any documentation explaining the meaning of the various flags shown in the Disk Report produced by the Acronis System Report tool.  E is understood to indicate that an error is detected but with no more suggestion as to what that error might be or even whether this is a serious issue or not?

It is possible that the Disk Report does not handle a RAID array scenario and is more intended for standard drive configurations but the only real way to understand this would be to raise it directly with Acronis by opening a support case to ask them about this.

I suspect that the Acronis first level support people will not be able to answer this question either, so you will need to press for it to be escalated to their expert team for an answer.

If you do go that route, then please do share back with us here in the forums, as am sure that lots of other users would appreciate having a clearer explanation!

I will guess,  the flags you speak of are shown under the Intel Raid 1 Volume 1.0 header.  My guess is the flags on both disks in the raid volume indicate the same error exists on both disks.

It is unlikely that both disks would have the same bad sector errors further, not all filesystem error flags will be reset by chkdsk /f /r. 

I suggest you run chkdsk c: /b

(the /b switch, NTFS only.  Clears the list of bad clusters on the volume and rescans all allocated and free clusters for errors).

Hello everyone,

We have a KB for disk report: https://kb.acronis.com/content/1638

In "more information" there is a "See also" box. I guess, this has the information that you were looking for.

Thanks Renata, that explains the codes nicely.

Martin,  your screenshot shows that errors exist on two partitions, number 1 and 2.  The number one partition looks to be a System Reserve partition.  That partition will not have a drive letter assigned to it.  To run chkdsk on that partition you will need to temporarily assign a drive letter to that partition  and then point chkdsk to that drive letter.  After chkdsk has been run you can remove the drive letter you assigned.

Replying to Renata

"We have a KB for disk report: https://kb.acronis.com/content/1638

In "more information" there is a "See also" box."

The 'more information' says the an 'E' means there is an error

No argument there ... We already know that

As to what causes the 'E' flag, the page, https://kb.acronis.com/content/1638,

That information page is silent

Replying to Enchantech

"to run chkdsk on that partition you will need to temporarily assign a drive letter to that partition  and then point chkdsk to that drive letter."

Thanks for that very good suggestion ... 

The result (as I expected) was no error

CHKDSK I: (System Reserved)

CHKDSK Results

Microsoft Windows [Version 10.0.18363.752]
(c) 2019 Microsoft Corporation. All rights reserved.

C:\WINDOWS\system32>chkdsk i:
The type of the file system is NTFS.
Volume label is System Reserved.

WARNING!  /F parameter not specified.
Running CHKDSK in read-only mode.

Stage 1: Examining basic file system structure ...
  256 file records processed.
File verification completed.
  0 large file records processed.
  0 bad file records processed.

Stage 2: Examining file name linkage ...
  364 index entries processed.
Index verification completed.
  0 unindexed files scanned.
  0 unindexed files recovered to lost and found.
  0 reparse records processed.
  0 reparse records processed.

Stage 3: Examining security descriptors ...
Security descriptor verification completed.
  54 data files processed.

Windows has scanned the file system and found no problems.
No further action is required.

    562175 KB total disk space.
     26544 KB in 97 files.
        88 KB in 56 indexes.
         0 KB in bad sectors.
      4983 KB in use by the system.
      4288 KB occupied by the log file.
    530560 KB available on disk.

      4096 bytes in each allocation unit.
    140543 total allocation units on disk.
    132640 allocation units available on disk.

C:\WINDOWS\system32>

I ran another Disk Report ... Sorry to say, same results ... 
 

I did some thinking about the Disk Report

This is an attempt to decode the column headings vs. the values in the columns

Any additional guesses will be appreciated

My observations/guesses

Decoding the Disk Report

Heading Value

Speed   ? 

IFace   Interface - seems access is through the RAID software (guess)

Hs-Bs-Tg    ? corresponding values are 0-0-0

Num     disk Number, also partition number

NT      Disk or partition

L9NO    What does this mean?

Size    HDD or Partition capacity

FSize   File System Size (guess)

Free    Free (unallocated) space

FS      File System (NONE or NTFS or free = not allocated to a partition)

Model   HDD Model Name
        SanDisk SDSSDRC032G
        Intel Raid 1 Volume 1.0. (actually NOT a HDD)
        
Type    Partition Type 
        from https://en.wikipedia.org/wiki/Partition_type
        73 - Reserved
        07 - MBR/EBR, LBA/CHS access - Windows NTFS
        27 - MBR, LBA access
    
ABCHSV  What does this mean?

(guesses)
A      Active
B      ?
C      ?
H      ?
S      ?
V      Volume

Martin,

Did you run chkdsk with the /b switch on all partitions on both drives in your raid 1 array?  If not you should.  Likewise of chkdsk /f has not been run on all partitions on both drives this should be done as well.

I have no idea what all of the L9N0 means nor the BCHSV,  A is for active as shown in the kb 1638 referenced by Renata shows.

Responding to Enchantech ... 

I am reluctant to run 'chkdsk /b' on a RAID array ...

If there were bad sectors on one disk in the RAID array, the same sectors would not be bad on the other disk in the mirror ... if chkdsk tried to correct that, the outcome might not be favorable

In any case, chkdsk /f /r checks both the file structure and checks all sectors on the partition (used and unused) no errors were reported by chkdsk /f /r

RAID verify has been run both on demand and on weekly schedule ... that tests every sector on each drive in the array and compares the data ... no errors have been reported

As indicated a few messages above, I assigned a drive letter to the System Reserved partition and ran a chkdsk ... no errors were reported

Martin, do you have a spare computer that would allow you to shutdown the RAID system, then take out each of the 2 disks one at a time and run drive diagnostics against them from the second computer?  That may be the only way to determine which drive in the array is generating this error.

Seagate disks: SeaTools for Windows would be able to check each drive without using CHKDSK.

Martin,

Steve has a good suggestion.  You have a raid 1 mirror array meaning that disk 2 of the array is a mirror image of disk 1.  If one of either disk in the array fail, the other disk can boot and work fine. 

I have a feeling that given your tests on disk 1 run clean that the errors are on disk 2.  If disk 2 has errors that cannot be corrected then that disk needs to be replaced.  If disk 1 fails and disk 2 has critical errors then you chance loosing your installation.

Bottom line is you need to figure out where the errors are, correct them if possible, and then hope that the fix holds.  If it doesn't then your looking at replacing the disk anyway.

How old are the drives in the array?  If they are older than about 5 years I would advise that you replace them both anyway, failures, errors, or not.

Responding to Steve Smith

I don't have another computer with SATA ports

Putting Drives from a RAID array into another computer  is problematic  

The BIOS RAID software will detect that the drives are part of a RAID array ... unexpected and unwanted side effects WILL occur ... I saw this happen when I installed a used drive to substitute for a failed drive in a RAID array

Responding to Enchantech

"I have a feeling that given your tests on disk 1 run clean that the errors are on disk 2."

Inter RST (Rapid Storage Technology) has a RAID Verify function

RAID verify and repair has been selected for about 1 month and is run weekly 

RAID Verify reports 0 media errors (0 bad blocks)

RAID Volume Data Verify and Repair

https://www.intel.com/content/www/us/en/support/articles/000006401/technologies.html

The drives are new -- both purchased on January 6 and 7, 2020

CHKDSK /f /r has been run 3 times and does not report lost clusters, does not report found files, does not report any file system errors, does not report any bad clusters

Chkdsk help info (partial)

CHKDSK [volume[[path]filename]]] [/F] [/V] [/R] [/X] [/I] [/C] [/L[:size]] [/B] [/scan] [/spotfix]

  volume              Specifies the drive letter (followed by a colon),
                      mount point, or volume name.
  filename            FAT/FAT32 only: Specifies the files to check for
                      fragmentation.
  /F                  Fixes errors on the disk.
  /V                  On FAT/FAT32: Displays the full path and name of every
                      file on the disk.
                      On NTFS: Displays cleanup messages if any.
  /R                  Locates bad sectors and recovers readable information
                      (implies /F, when /scan not specified).

If we have to choose between a mysterious error that neither chkdsk nor Intel RST can detect on the one hand and the possibility that the Acronis Disk Report is giving a FALSE error indication ... Occam's razor tells us that the Disk Report is misleading us

Occam's razor https://en.wikipedia.org/wiki/Occam%27s_razor

Martin, I wouldn't connect your SATA drives directly to a SATA port inside another computer but would connect via a USB > SATA adapter or dock for any testing.  Not sure if there would be any side effects of using that method as I do not have any RAID systems myself (other than single drives configured to use RAID by the vendor).

Responding to Steve Smith

"Martin, I wouldn't connect your SATA drives directly to a SATA port inside another computer but would connect via a USB > SATA adapter or dock for any testing.  Not sure if there would be any side effects of using that method as I do not have any RAID systems myself (other than single drives configured to use RAID by the vendor)."

Seagate SeaTools does NOT work when using a SATA to USB bridge 

I have tried this with a SATA/IDE to USB bridge -- SeaTools does NOT recognize the HDD

For some 2.5 inch notebook HDD I needed to test, I purchased this 

Sabrent 2.5-Inch SATA to USB 3.0 Tool-Free External Hard Drive Enclosure

https://smile.amazon.com/gp/product/B00OJ3UJ2S/

It is nice looking, no-tools install ... SeaTools recognizes the drive in that configuration, but 2 different drives FAIL SeaTools testing -- even though those drives seem to operate properly -- Format OK, chkdsk OK

Also someone else had similar experience

SATA HDD not being seen using a USB bridge

https://forums.tomshardware.com/threads/sata-hdd-not-being-seen-using-a-usb-bridge.1758769/

Martin,

Thanks for the additional info on your system.  I agree with you that sense both chkdsk /f/r and the Intel verify function both show no errors things are fine. 

Since the E flag does not appear to cause any issue with the backup of your raid array then I think it safe to ignore the flag.  Not sure when the Acronis Disk Report tool was released so it might not be compatible with recent changes in Intel RST controller driver software which causes the error.

Responding to Enchantech, 

"Since the E flag does not appear to cause any issue with the backup of your raid array then I think it safe to ignore the flag."

If only it was that simple ... 

I have a case open via Acronis chat ... I have collected 4.02GB of information (scans, reports, screenshots, commentary), including 21 chat transcripts ... over a 3 week period.

The chat agents will not escalate my case to the Expert Team until I can submit a Disk Report without the 'E' flag

Okay, so what exactly is the issue that you have an open case with support about?

Responding to Enchantech

There are 3 cases

2020 02 27 my NAS ran out of space; all of the backups (TIBX) for that computer were corrupted after that out of space occurrence Those backups had to be abandoned (deleted) and a new backup configuration had to be  established

2020 03 27 Backup performance issues; backup speeds were both fast and painfully slow; Full backup speeds were 407mbps; 35mbps, 207mbps, and 141mbps ... backup durations ranged between 3 hours 23 minutes and 26 hours (for those, backup size was between 407GB and 416GB) ... Incremental backup speeds were between 3.2mbps and 25mbps. I was encouraged to run a backup verification -- that ran for 102 hours -- that is 4 DAYS and 6 HOURS

2020 04 01 Acronis reported a network share failure -- unexplained.

As part of that ongoing investigation, I configured a similar backup of a notebook PC (also on the same wired LAN) Backup speeds were 239mbps (full, 25GB), 25mbps and 804kbps (Inc, under 1GB) then 291mbps (full 26GB) and 4mbps (Inc 144MB) ... Under 1 hour for 5 backups and about 53GB)

In addition, I made a new backup configuration for my desktop PC ... backup speeds were also reasonable ... the slow backup speeds did not occur for that configuration ... Since 4/1 I have tried to have the slow  and erratic backup speed of the prior backup configuration explained ...

Over 20 chat sessions ... they always find something else that needs to be answered before they can escalate the case to the Expert Team ... The 'E' flag has been a persistent stumbling block -- sometimes I get an agent to agree it does not apply but with a different agent in a new chat it becomes 'important' again ...

chkdsk was another ongoing issue -- one said to run chkdsk /f /r and another agent said I had to run chkdsk /r /f another tried to claim that the chkdsk flags had to be upper case ... then (after it had been run) they had to see a screen shot -- even though chkdsk run at boot time does not give a useful screenshot ... one agent said that kind of chkdsk had no accessible log -- That is false, actually it is available in Event Viewer ... when I showed another agent the chkdsk I had copied from Event Viewer, He insisted I show a screenshot of running chkdsk from the command line ... I had to show him a screen shot of chkdsk from a command prompt with the message that it will be run at the next boot

VSS has been another on/off issue

My LAN speed has been an issue ... I can prove it is gigabit capable, they still doubt

When I mention a speedtest to verify my LAN speed they go off on a tangent about backups to the Acronis cloud -- even though we have been talking about backups to my NAS during that chat

It is like I am training them to do their job 

I am still waiting to find out why the some backup configurations run with such erratic speed

I don't have confidence when a backup can run in 2 1/2 hours one day and might take more than a day another instance ... I don't like it when a chat agent tells me backup time are unpredictable

Since 1958 I have been a computer programmer, System Architect, I have created database apps from scratch (not using a database package), I have created computer languages (language definition, compiler, and run time environment) ... I don't like being treated like the village idiot by someone who doesn't know how to enter a chkdsk command and get the results

Martin,

Thank you for replying with your case information.  Your experience with Support leaves much to be desired to say the least.  You certainly have the background to understand and troubleshoot issues with your computers.  The sad fact is (you probably already know this) that front line support read from prepared scripts for troubleshooting.  So when things get beyond the common user mistake or a known issue, things so south quickly.

I may not have the answers your looking for but I can provide you some reasoning as to the behavior you have experienced.  Hopefully, you may find something here that makes a difference.

2020 02 27 my NAS ran out of space; all of the backups (TIBX) for that computer were corrupted after that out of space occurrence Those backups had to be abandoned (deleted) and a new backup configuration had to be  established

The issue here was as you say, an out of space problem with your NAS.  A contributing factor is your use of the incremental backup method.  It is a known documented issue that incremental method backups are dependent on each other which means of course that if one incremental is corrupt the entire backup task is useless.  This is not a completely new problem.  What is new is that the new TIBX format for backup files requires that an incremental method backup task has all subsequent backup files created by the backup task incorporated into the first Full backup created by the task.  This is a change to the application that has been the Achilles heal of 2020.  Users whom choose to use the incremental method, which is also ironically the default method, did not by and large know of this change in the apps behavior.  That fact unfortunately brought many, many users issues like that of yourself with using the incremental method.

My advice to those whom wish to use the incremental method is that they develop a backup plan based on total available storage capacity that allows for an initial full selected data backup followed by a limited small number of incremental backups (think 5 to 7) followed by a repeat of that scheme.  Additionally you must establish clean up rules that will not overrun your available storage capacity nor create long chains of backup files, (think 3 or 4 sets).  Having said that if you have a lot of data that changes in big ways in short time periods these recommendations need to be adjusted accordingly or changing to a differential method should be considered.  Differential method backups are not interdependent like incremental backups.

NOTE:  The above applies to new full disk backup tasks created by TI 2020.  Folder/file backups created by TI 2020 still use the TIB format where incremental backup files dependency is tied only to a single full backup they are based upon which allows for longer chains to be created without loosing the entire backup chain.

Essentially the difference here is that with the TIBX format the meaning of Version Chain has changed with respect to full disk backups.  A version chain now includes all backup files created by a backup task regardless of any subsequent full version(s) within the backup scheme.

2020 03 27 Backup performance issues; backup speeds were both fast and painfully slow; Full backup speeds were 407mbps; 35mbps, 207mbps, and 141mbps ... backup durations ranged between 3 hours 23 minutes and 26 hours (for those, backup size was between 407GB and 416GB) ... Incremental backup speeds were between 3.2mbps and 25mbps. I was encouraged to run a backup verification -- that ran for 102 hours -- that is 4 DAYS and 6 HOURS

This behavior is the result of yet another change in TI 2020 which has impacted users at large as well.  At issue here is the use of metadata tracking of the data set configured in the task.  In your example of backup totals of 407GB and 416GB would mean a large amount of metadata generated for tracking purposes.  The reasoning for use of metadata tracking is to increase reliability of backups.  Full disk backups are performed at the block level on disk.  Therefore it is logical to assume that metadata is created at the block level.  Standard allocation size for an NTFS volume is 4096 bytes so metadata for each 4096 byte data segment is generated.

Knowing this lets consider the usage of an incremental method with respect to the above.  It should be obvious that to create an incremental backup based on an over 400GB full file means that all the metadata tracking must be examined and compared with the data snapshot created by the application.  This of course takes time and is somewhat dependent on individual machine hardware (total available ram, disk interface and disk type).  After the comparison completes then the backup task begins to create the backup file.  Again, new metadata tracking is compiled for this incremental file.  Now, imagine this repeating for each new incremental and full file within the backup scheme version chain.  The longer the chain the longer the backup takes.  The dependency of incremental method backups then creates an undesirable time frame then for large total amounts of data.

2020 04 01 Acronis reported a network share failure -- unexplained.

As part of that ongoing investigation, I configured a similar backup of a notebook PC (also on the same wired LAN) Backup speeds were 239mbps (full, 25GB), 25mbps and 804kbps (Inc, under 1GB) then 291mbps (full 26GB) and 4mbps (Inc 144MB) ... Under 1 hour for 5 backups and about 53GB)

This behavior is something I cannot explain.  I witness it too.  I suspect that it is a result of throttling by the application when the amount of data being backup is small in total size.  Again, the use of the incremental method is playing a role in your results.

In conclusion it is my assessment that TI 2020 changes in backup operations with respect to the TIBX format and full disk backups has made it necessary that user reevaluate their backup methods and in turn may well dictate big changes in the backup plan of the user.  The trade off for that is that backup file reliability has improved as a result.  To that end the user must be mindful of the implications that the TIBX format brings in order to achieve improved reliability and acceptable performance levels.

Responding to Enchantech

Thank you for your attention to detail, understanding, and advice

Right now my NAS has 3 folders of backup data

The slow (failing) backup -- 1.03TB (marked as do not schedule)

The 'test' backup -- 885GB  (marked as do not schedule)

Backup of another PC -- 600GB (estimated)

NAS available space after considering the above requirements is on the order of just 200GB

Currently the NAS does not have enough space to resume any backups of my desktop PC

Acronis chat agents have asked that I refrain from deleting any backups -- as they may be needed by the Expert Team ... Fat chance that will happen since they will not escalate the issue

I am considering going against their wishes, delete both backups to free space on my NAS and creating a new backup configuration -- Full plus 5 incrementals, store no more than 2 chains will work and not exceed available space ... As an alternative, Full plus 5 differentials no more than 2 chains

I am also considering rolling back to Acronis 2019 so that the backups will be in the TIB format instead of the TIBX format

I know that after the backup configuration is setup for TIB format I could reinstall Acronis 2020 and backups will continue to be in the TIB format

I have never used Acronis differential backup ... I will know neither the storage requirements not the backup speed until I implement it

Do you have any comments or suggestions?

Martin,

Just speaking for myself here.  I would nuke your current backups and start again.  Your plan of backup I believe to be sound.  I think from a performance standpoint a differential method will outperform that of an incremental.  Having said that, I have not tried the incremental method since the newest 25700 version was released so I could be surprised!  I have an incremental method backup task configured from last year to an NAS device.  Out of curiosity I have turned on the schedule to see how it will work. This will be a weekly backup full + 5 inc. no more that 3 versions.  We'll see!

When Ti 2020 was released I moved completely to the tibx format.  I also moved to differential backup methods.  I am pleased with the results thus far.  As for performance these tasks  remain consistent from the start.  Time, throughput, are within seconds and a few MB of each other from backup to backup.

That's my 2 cents worth.

I have some news ... 

1. I opened a chat session and found that instead of a promised phone call, they sent an eMail that requested yet another report -- WireShark packet log while a backup was running

2. I pointed out that I don't have enough remaining space on my NAS and that I was told not to cleanup any backups as that might be needed for this ongoing investigation

3. Guidance was that I should cleanup the oldest ... looking at the cleanup  data, I found that the failing (slow) backup configuration had incremental backups of about 800GB ... even though disk space in use is about 680GB

4. The chat agent asked me to run the failing backup with an advanced option 'snapshot -> none'

5. Conclusion ... VSS can't create a snapshot when more than 50% of the HDD is allocated

6. Ran a backup configuration ... 35GB was backed up in 1 hour 6 minutes ... speed was 72.7Mbps ... ACCEPTABLE

7. All of the other troubleshooting, wild guesses, false paths about what could have been causing the slow backups: NAS disk speed, LAN speed, the 'E' flag, CHKDSK ... ALL of that was a Red Herring
 

Martin,

Glad to see you found resolution however, I'm not sure I buy the "5. Conclusion ... VSS can't create a snapshot when more than 50% of the HDD is allocated"

VSS should never require more than 15% of free disk space for a snapshot.

If your target disk had less than 10 percent free space (per Microsoft) when the backup task was executed then VSS could fail and the backup would fail then as well.  I have no idea where the 50% figure came from but that is way out of line.

Responding to Enchantech

I initiated enough backups (incremental) so that Acronis would run a full backup 

The initial incremental backup(s) had reasonable time(s)

But the full backup ran for almost 7 hours ... contrast that with an initial full backup that I have seen take about 2.5 hours

BTW, that 7 hour full backup is better timing than the previous full backup in that configuration which took 41 HOURS --- that is 1 DAY and 17 HOURS

At this point, I will run the packet monitor and submit for analysis

Martin,

Looking at your screenshots I have some questions, hopefully you won't mind answering.

1. Can you verify that the screenshots show a continuation of an existing task and not a newly configured task?
2. On April 4 a new Full version was created which consolidated a total of 463.9GB out of a total of 527.6GB of data.  That task ran at a miserably slow 28.8Mbps.  Can you provide details of your network and your NAS?  Is the NAS on a wired connection? What are the Specs of the NAS?  What are the Specs of the PC used as source?
3. On April 10 another Full backup was created and consolidated 483.4GB out of a total of 534.2GB of data.  That task ran at more respectable 135.9Mbps but still not what I would expect to see for a full backup of this much data.  Can you tell me what the total data consists of?  Are the majority of the data being backed up already compressed files such at photos, music, videos etc.?
4. During the time of the April 4 backup was there any other traffic on your network such as downloading online game play or streaming?  On the April 10 backup was there other traffic on your network and if so what?

Network backup performance is highly hardware dependent.  Example:  I have two NAS devices on my network.  One of them is a consumer grade 4 bay device having an Intel 2 core Celeron J1800 CPU @ 2.4Ghz, 2GB of RAM, and 4 - 6TB Seagate Iron Wolf NAS 7200rpm drives.  This device routinely backs up over my wired 1Gbps network at around 450Mbps.

My second device is a full fledged Linux based server sporting 4 drives - WD Red 3TB 7200rpm having an Intel Quad core i5 CPU @ 3.2Ghz, 16GB RAM.  This device routinely backs up at around 500Mbps.

I am curious if your hardware or network configuration may be playing a part in your poor performance.

Responding to Encantech

"1. Can you verify that the screenshots show a continuation of an existing task and not a newly configured task?"

An existing backup configuration that has been idle since I created the test2 configuration ... I didn't want to be running 2 'competing' backups of the same source

"2. On April 4 a new Full version was created which consolidated a total of 463.9GB out of a total of 527.6GB of data.  That task ran at a miserably slow 28.8Mbps.  Can you provide details of your network and your NAS?  Is the NAS on a wired connection? What are the Specs of the NAS?  What are the Specs of the PC used as source?"

I have a text file with that information

"3. On April 10 another Full backup was created and consolidated 483.4GB out of a total of 534.2GB of data.  That task ran at more respectable 135.9Mbps but still not what I would expect to see for a full backup of this much data.  Can you tell me what the total data consists of?  Are the majority of the data being backed up already compressed files such at photos, music, videos etc.?"

Not sure ... probably mostly compressed (music, pictures, executables)

Whatever tha content, it remains consistent day by day by day

"4. During the time of the April 4 backup was there any other traffic on your network such as downloading online game play or streaming?  On the April 10 backup was there other traffic on your network and if so what?"

April 4, 10 I doubt there was an streaming

Certainly no streaming at the same time as the 4/25 full backup that started at 1 minute after midnight

"I am curious if your hardware or network configuration may be playing a part in your poor performance."

Doubtful -- In Windows Task Manager, I see LAN upload rates as high as 600Mbps

BTW, watching the backup ... At first there was a burst of activity ... perhaps 5 minutes ... then an 20+ minute time with low usage by Acronis and no LAN activity

In addition, fast.com confirms that my access to the internet is 82Mbps (down) over 97Mbps (up) ... although that does not confirm the gigabit capability of my LAN, if there was an issue with my LAN, I wouldn't be able to see that kind of speed

Just now, I uploaded 7.5GB of music (English Renaissance) ro a folder on my NAS ... took about 3 minurtes and Task Manager shows over 500Mbps upload speed

Thank you for your interest and attention to detail

--MGD

Please Stay Safe

Martin said:

"3. On April 10 another Full backup was created and consolidated 483.4GB out of a total of 534.2GB of data.  That task ran at more respectable 135.9Mbps but still not what I would expect to see for a full backup of this much data.  Can you tell me what the total data consists of?  Are the majority of the data being backed up already compressed files such at photos, music, videos etc.?"

Not sure ... probably mostly compressed (music, pictures, executables)

Whatever that content, it remains consistent day by day by day

Can you take a look at the Backup tab in the TI GUI, toward the bottom of the right hand window.  You will see a legend that reads: Pictures – Videos – Audio – Documents – System – Others.  This is a breakdown of what TI found on disk for the backup. Can you post what these categories show?

You state that content on this disk is consistent day by day by day.  Please explain then how it is that beginning on:

• Yesterday at 2:14PM the task created a backup of 35.1GB
• Yesterday at 6:57PM the task created a backup of 9.1GB
• Yesterday at 9:35PM the task created a backup of 2.1GB

46.3GB of changes is a lot of data, can you explain this?

Martin said:

"4. During the time of the April 4 backup was there any other traffic on your network such as downloading online game play or streaming?  On the April 10 backup was there other traffic on your network and if so what?"

April 4, 10 I doubt there was any streaming

Certainly no streaming at the same time as the 4/25 full backup that started at 1 minute after midnight

Were there any mundane network activity like web surfing during these time periods?

Martin said:

 "I am curious if your hardware or network configuration may be playing a part in your poor performance."

Doubtful -- In Windows Task Manager, I see LAN upload rates as high as 600Mbps

BTW, watching the backup ... At first there was a burst of activity ... perhaps 5 minutes ... then an 20+ minute time with low usage by Acronis and no LAN activity

Figures shown in Task Manager can give you a feel for performance however they can also be misleading.  Running a backup task has many parts that effect performance, compression, varying file sizes, and other factors play into performance.

Your statement “At first there was a burst of activity ... perhaps 5 minutes ... then an 20+ minute time with low usage by Acronis and no LAN activity”,   I find concerning however.  This suggests issues with your network, and you should not be experiencing them.

Martin said:

In addition, fast.com confirms that my access to the internet is 82Mbps (down) over 97Mbps (up) ... although that does not confirm the gigabit capability of my LAN, if there was an issue with my LAN, I wouldn't be able to see that kind of speed

Just now, I uploaded 7.5GB of music (English Renaissance) ro a folder on my NAS ... took about 3 minutes and Task Manager shows over 500Mbps upload speed

Internet speed sites are not good comparisons for LAN speeds, an apples to oranges comparison really.

Your 7.5GB transfer averages to just 357.9Mbps.

I think your network is where your performance lag is taking place.  I note that you have a FIOS router and Dlink switch.  When you list how your devices are connected it sounds like the only device connected to the switch is your notebook.  The rest of your devices then I assume are connected to the FIOS router, Why?

The Dlink switch is far more capable in switching duties than the FIOS router.  I think you need to reconfigure your network so that all your devices, desktop, notebook, NAS, etc., are all connected to the Dlink switch and the Dlink alone is connected to the FIOS router.  I would bet that this one change would solve the majority of your issues.

Another performance bottleneck that I see is that you only have a single 5400 rpm drive in your NAS and that NAS only has 512MB of RAM.  Your performance would double with the addition of another 1.5 gigs of ram and a second HDD.

I am attaching two screenshots here of a task I created that I hope will give you a comparison between where you are now and what is possible.  Refer to my specs I mentioned earlier for differences in hardware of my NAS devices.

I do not encourage that you or any other user reading this create a task like I have pictured.  It is a one-off task and it and the backups it creates will be deleted from my system when I have finished with it.

Screenshot one is that of the Activity tab view of the task.  You will see 4 backups listed.  The top two are backups of a 2TB storage disk containing 549GB of data.  The top backup I ran to my FreeNAS server.  The second backup I ran to my 4 bay NAS device.  Notice that performance is better to the NAS device than to the server.  This fact is also true for the next two backups shown as well.  Note that the second backup shows as a differential.  That is because I set the task up as a differential method task.  Also notice that backup file size differs from the Server (larger) and the NAS (smaller).  That can be explained in the difference in filesystems used on these NAS devices.

The last two backups are images of my Win 10 OS drive, a 512GB M.2 NVMe PCIe 3.0 drive.  Again, the backups were run to my server and my NAS respectively.  Note that as data size increases so does average transfer speed not by huge amounts but there certainly are increases.

Make the changes I suggested and then give things a try again.  I believe you will see some improvement.

Note in the screenshot below the legend of contents of the backup.  Pics, Video, Audio are already compressed data and do not compress further even though TI tries to do so.  If most of your data is of this type you may wish to consider not using compression which will increase performance.

Responding to Enahantech, 

"I think your network is where your performance lag is taking place.  I note that you have a FIOS router and Dlink switch.  When you list how your devices are connected it sounds like the only device connected to the switch is your notebook.  The rest of your devices then I assume are connected to the FIOS router, Why?

The Dlink switch is far more capable in switching duties than the FIOS router.  I think you need to reconfigure your network so that all your devices, desktop, notebook, NAS, etc., are all connected to the Dlink switch and the Dlink alone is connected to the FIOS router.  I would bet that this one change would solve the majority of your issues.

Another performance bottleneck that I see is that you only have a single 5400 rpm drive in your NAS and that NAS only has 512MB of RAM.  Your performance would double with the addition of another 1.5 gigs of ram and a second HDD."

The 8-port D-Link switch has ports in use that connect to other devices in the house

4 ports on the FIOS router were in use ... Synology, desktop PC, another PC, and the D-Link

As far as getting off the FIOS router ... I was considering doing that ("Great minds think alike") I have a Tenda 5 port Gigabit switch ... I moved the 4 cables from the FIOS router to the Tenda and now have just 1 LAN patch cord into the FIOS router

I made this change to remove a vulnerability -- I realized that the FIOS router could get reset by Verizon -- which could disrupt a backup on my LAN

"Just now, I uploaded 7.5GB of music (English Renaissance) ro a folder on my NAS ... took about 3 minutes and Task Manager shows over 500Mbps upload speed

Internet speed sites are not good comparisons for LAN speeds, an apples to oranges comparison really.

Your 7.5GB transfer averages to just 357.9Mbps."

I reran that uplink test twice and took an accurate note of the transfer time 2:20 and 2:28 ... the progress window showed about 48MB/s to 65MB/s speed -- that would be about 400 Mbps to 520Mbps

Martin,

Good, now test out that backup task and see if things improve.

Responding to Enchantech

After the rewire of the LAN drops, no change in the backup timing

The last 4 incremental backus have each taken about 30 minutes ...

I suspect this is the amount of time needed for Acronis to scan my local C: drive and the metadata

I have 548GB allocated; 546,707 files; 84,023 folders

For a home/office setup, would that be an unusually large file system?

BTW, (clarification) My Synology NAS has 2 3TB drives in a RAID 1 configuration

--MGD

Please Stay Safe

P.S. The folder where I have all of the troubleshooting information for the current active case occupies 11GB, 142 files (my notes, screenshots, environment description, and 23 saved transcripts)

Martin,

So no improvement is noted after reconfiguration of the topology.  So this then points to machine hardware being the next possible source of trouble.  

An easy step would be to watch your DS216 Memory Monitor during an ATI backup session.  If you witness high memory usage then this indicates that more ram would benefit the NAS.  I have read that the DS216 memory is soldered to the motherboard and therefore cannot be changed yet I read that the DS216+ memory can be upgraded.  Best to check on that yourself.

Using a 2 disk raid 1 mirror will net you single disk performance.  I think WD Red 3TB 5400rpm drives are rated at around 112MBps on an internal 6Gbps SATA connection.  Of course you will not get that much over Ethernet. 

You do not say how many GB of data these last 4 backups.  I assume that your previously posted screenshot of the Activity tab showing 3 incremental backups are 3 of these 4 so working with that if we take the 9.6GB backup which showed an average transfer rate of 48.7Mbps, in time that calculates to slightly over 25 minutes.  The activity shows time elapsed of 27 minutes and 13 seconds.  That would mean approx. 2 min and 13 seconds for snapshot creation.  The 48.7Mbps equates to a miserable 6MBps.  That suggests that hardware between your 3TB drives and your source PC is creating a bottleneck. 

Did you notice the period of inactivity during this last backup that you mentioned before?

I would say keep looking.  You just haven't found the issue yet.

Responding to Enchantech

"watch your DS216 Memory Monitor during an ATI backup session."

Around 50% memory usage during a backup ... probably not an issue

My Synology is a DS216j ... the datasheet does not indicate upgradable RAM 

"You do not say how many GB of data these last 4 backups"

New screenshot included here

"period of inactivity"

The scheduled backup is at 1 minute after midnight ... Between staying isolated at home; keeping up with Facebook, and household tasks ... I am tired enough each day to keep a regular wake/sleep schedule -- at midnight, I just let the PC and the Synology talk to each other, but I don't watch

More seriously, on a manual backup, or one that intrudes on my daily activities, I watch like a hawk, screen shots, Task Manager, whatever might give me a hint of a cause/solution

"I would say keep looking.  You just haven't found the issue yet."

Did  I mention that I have been a computer programmer since 1958? ... Looking for hardware /software bugs is like breathing for me

--MGD

Please Stay Safe

Martin,

It just might be that your NAS having a 2 disk raid 1 setup is the bottleneck.  Obviously ram is sufficient.  I know that on my NAS device I will only see about 40% memory usage during a backup.  On my server that is usually around 20% memory usage.  I know that your Synology device uses a EXT3 filesystem which is a good performer however it is not a next generation filesystem that handles multiple volumes well like XFS or ZFS filesystems.  My NAS device uses the Btrfs filesystem which has its shortcomings like EXT4 but is otherwise a good performer. 

Have you looked at CPU usage on your NAS during a backup?  I know on mine I usually see about 45% with spikes as high as 90% that are usually short burst.  These spikes do drag on throughput though so worth having a look.

Responding to Enchantech, 

"It just might be that your NAS having a 2 disk raid 1 setup is the bottleneck."

Please consider that the test2 backup configuration consistently performed much better than the failing B667REBUILD4E backup configuration ... Both of those Acronis backup configurations use the same source HDD, same LAN, same Synology NAS

The next scheduled backup will be a full backup ... tomorrow we will know more

--MGD

Please Stay Safe

Martin,

I'm sorry but I cannot locate any screenshot you have posted that shows the test 2 backup nor can I find reference to it in any posts to this thread.  Can you please post or re-post this?

Responding to Enchantech, 

"I'm sorry but I cannot locate any screenshot you have posted that shows the test 2 backup nor can I find reference to it in any posts to this thread.  Can you please post or re-post this?"

My bad -- sorry ... Also included screenshots of the troublesome backup

I'm reminded of some office humor I read while I was consulting at IBM in Poughkeepsie ... 

"Can console logs be used as evidence?"

--MGD

Please Stay Safe

Responding to Enchantech, 

Here is a screenshot of the latest backup activity - it took 4 hours, 50 minutes to backup 420GB

I also included 2 additional screenshots to show the initial backup for the current configuration and the initial run of the test2  configuration

Contrast the latest backup with the initial backup which took 3 hours, 23 minutes to backup 407GB

Contrast the latest backup with the initial test2 backup which took 2 hours, 39 minutes to backup 404GB

Latest to best is 1.82 times as much time for only 1.04 times as much content

This doesn't seem to be working properly yet ... and still no clue as to why the wild variation in backup speed

--MGD

Please Stay Safe

Martin,

Martin said:

Contrast the latest backup with the initial backup which took 3 hours, 23 minutes to backup 407GB

Compared to the latest backup of 420GB taking 4 hours and 50 minutes, note that a cleanup of the task occurred with the latest backup in addition to the added 13GB of data.  Those two factors are at play here and attribute to the time difference.

Martin said:

Contrast the latest backup with the initial test2 backup which took 2 hours, 39 minutes to backup 404GB

Compared to the above two backups again involves the cleanup and added data. 

The best comparison here will be the next backup of the B667REBUILD4E task.  After the cleanup happens what will the performance be for the next incremental vs. the first incremental of the test2 task allowing for any difference in total data backed up between the two tasks.

During the run of the latest 420GB backup did you watch what happened during the backup?  If so did you notice any periods of inactivity or slowness?  If you did how long did such periods last approximately? 

If your backup is comprised of mostly already compressed data how about running a test in which you turn off data compression.  I think you might be surprised at the results.

In all honesty, I have given up on the incremental scheme.  I find the differential scheme to be more efficient and better performing.  Your mileage may vary.

Responding to Enchantech, 

"The best comparison here will be the next backup of the B667REBUILD4E task.  After the cleanup happens what will the performance be for the next incremental vs. the first incremental of the test2 task allowing for any difference in total data backed up between the two tasks."

OK ... I will continue to monitor and report

"During the run of the latest 420GB backup did you watch what happened during the backup?  If so did you notice any periods of inactivity or slowness?"

I did not observe ... at midnight

"If your backup is comprised of mostly already compressed data how about running a test in which you turn off data compression."

I checked Backup -> Options -> Advanced -> Performance

The option for no compression is 'greyed-out' -- can't be changed

That's all until tomorrow

--MGD

Please Stay Safe

Martin,

Just so you know, to enable the "no compression" option you will need to configure a new task.  Once a configured task is run the compression option cannot be changed.

Responding to Enchantech, 

"to enable the "no compression" option you will need to configure a new task."

Thanks

:The best comparison here will be the next backup of the B667REBUILD4E task.  After the cleanup happens what will the performance be for the next incremental vs. the first incremental of the test2 task "

Very good prediction ... 

B667REBUILD4E 10.4 GB  41.3Mbps  34m 30s  541.2GB

test2          6.1GB   43.1Mbps  19m 29s  527.5GB

In other words, "I have some good news and some bad news" ...

The good news -- It seems to have 'fixed' itself...

The bad news -- We may never know why

--MGD

Please Stay Safe

Martin,

After seeing your latest result I believe things are working as they should.  You may never know what the root cause of the issue was true, I would say though that reconfiguration of your topology helped.  Another factor may have been an unseated Ethernet cable.  I have experienced this sort of thing before where a cable becomes unseated by a small fraction causing signal drops, speed drops, etc.. 

:-)

Responding to Enchantech, 

"I would say though that reconfiguration of your topology helped."

Doubtful as I was always able to confirm gigabit speed throughout the month long investigation

"Another factor may have been an unseated Ethernet cable."

That could happen when the RJ-45 locking tab is missing ... All of my cables are intact.

I work as a field tech ... one of the things I expect to find during a work assignment

--MGD

Please Stay Safe

Martin,

I wouldn't be as sure as you are with cabling.  I just had an associate several days ago having issues with network speed and found that disconnecting and the reconnecting the Ethernet cable to the NIC restored full Gig speed.  I my self have had more than one occasion where cables appear good, not damaged yet, issues persist, slow speeds, dropped connections, etc.

My issue with Disk Report has been escalated

I have been getting phone calls about this issue and an eMail inquiry was made to the developers to identify what conditions set the 'E' flag on the Disk Report

I'm 'excited'

I expect more information ... sometime this year

--MGD

Please Stay Safe

Greetings, 

I never did get an explanation for the 'E' flag -- Still waiting for about 20 months ... SMH!

--MGD

Please Stay Safe