Copying timestamps among Linux systems

Collapse
X
Collapse
 
  • Page of 2
  • Time
  • Show
Clear All
new posts
Previous 1 2 template Next
  • Carsten
    Journeyman
    • Oct 2015
    • 18
    #1

    Copying timestamps among Linux systems

    Hello all,

    I'm using BC 4.2.4 on Ubuntu 18.04 LTS, newly installed , comparing a local folder to a remote folder via SFTP (SSH2). The remote system is Ubuntu 16.04 LTS.

    On the remote system, I have a file with this timestamp:

    Code:
    carsten@ProLiant-Ubuntu:~/LaufwerkD$ stat Fliegerei/Halle/Hörmann/Hörmann\ RAL-Karte.pdf 
  Datei: 'Fliegerei/Halle/Hörmann/Hörmann RAL-Karte.pdf'
  Größe: 1549888   	Blöcke: 3032       EA Block: 4096   Normale Datei
Gerät: 801h/2049d	Inode: 67766388    Verknüpfungen: 1
Zugriff: (0664/-rw-rw-r--)  Uid: ( 1000/ carsten)   Gid: ( 1000/ carsten)
Zugriff    : 2018-04-27 20:25:38.838392907 +0200
Modifiziert: 2018-03-14 11:19:22.000000000 +0100      # <----- timestamp of remote file
Geändert   : 2018-03-25 12:15:28.647500113 +0200
 Geburt    : -
carsten@ProLiant-Ubuntu:~/LaufwerkD$ stat Fliegerei/Halle/Hörmann/Hörmann\ RAL-Karte.pdf --format %Y
1521022762
    When I sync/copy this file to the local system with BC, I get this timestamp:

    Code:
    carsten@red-glow:~/LaufwerkD$ stat Fliegerei/Halle/Hörmann/Hörmann\ RAL-Karte.pdf
  Datei: Fliegerei/Halle/Hörmann/Hörmann RAL-Karte.pdf
  Größe: 1549888   	Blöcke: 3032       EA Block: 4096   Normale Datei
Gerät: fd01h/64769d	Inode: 47318200    Verknüpfungen: 1
Zugriff: (0664/-rw-rw-r--)  Uid: ( 1000/ carsten)   Gid: ( 1000/ carsten)
Zugriff    : 2018-03-14 11:19:20.000000000 +0100
Modifiziert: 2018-03-14 11:19:20.000000000 +0100      # <----- timestamp of local file, copied with BC
Geändert   : 2018-04-27 20:29:47.758882659 +0200
 Geburt    : -
carsten@red-glow:~/LaufwerkD$ stat Fliegerei/Halle/Hörmann/Hörmann\ RAL-Karte.pdf --format %Y
1521022760
    That is, the local file is 2 seconds older.

    Why is that and what can I do about this?

    If I copy the same file with
    Code:
    scp -p ...
    the timestamps are identical, so the issue does not seem to be related to one of the involved systems.
    The same problem also occurs with other files whose path does not contain spaces or special characters.
    It is reproducible, but only with very few files, roughly one in twenty or so.

    Best regards,
    Carsten
    Last edited by Carsten; 27-Apr-2018, 02:38 PM. Reason: Added details about path names and reproducibility.
    Tags: None
    • Aaron
      Team Scooter
      • Oct 2007
      • 15997
      #2
      Hello,

      What file systems are you working with on the source or destination? A FAT32 transfer can introduce up to a 2 second difference (which can optionally be ignored as within tolerance by Beyond Compare).
      Aaron P Scooter Software

        Comment

        • Carsten
          Journeyman
          • Oct 2015
          • 18
          #3
          Hello Aaron,
          thanks for your reply!

          Originally posted by Aaron
          What file systems are you working with on the source or destination? A FAT32 transfer can introduce up to a 2 second difference (which can optionally be ignored as within tolerance by Beyond Compare).
          The file was originally copied to the source system from Windows from an NTFS partition, using BC4 for Windows. I guess this is how it got the timestamp of 11:19:22.000000000 (integer seconds) on the source system, whose file system is ext4, Ubuntu 16.04 LTS.

          The target's file system is ext4 as well, using Ubuntu 18.04 LTS.

          Thus, it looks to me as if the NTFS roots of the file cannot not play a role except as the origin of the integer value for the seconds. The actual copy I'm worried about is between ext4 on both the source and the destination, which I had expected to be accurate, as is the case with most other files and when the file is alternatively copied with scp -p

            Comment

            • Aaron
              Team Scooter
              • Oct 2007
              • 15997
              #4
              Hello,

              Copying with SCP is a secure copy but is not equal to an SFTP transfer. Different transfer protocols can round the timestamps differently when handling different filesystems. For an alternate test, you could try other SFTP clients, such as Filezilla, to see how they behave.

              If you recopy one of the trouble files (the 1 out of 20), does it recopy with the same timestamp every time? If you delete the destination file (to avoid overwriting) first, and then re-copy: what timestamps does this get?

              One manual override is the Touch command, which can manually set the timestamp (after you verify the files are binary equal with a Compare Contents command).
              Aaron P Scooter Software

                Comment

                • Carsten
                  Journeyman
                  • Oct 2015
                  • 18
                  #5
                  Hi Aaron,

                  Originally posted by Aaron
                  Copying with SCP is a secure copy but is not equal to an SFTP transfer. Different transfer protocols can round the timestamps differently when handling different filesystems. For an alternate test, you could try other SFTP clients, such as Filezilla, to see how they behave.
                  I just tried this with FileZilla, using an SFTP connection: Just as scp, it copies the timestamps properly, resulting in "… :22". It's only BC4 that has an offset of 2 seconds, yielding "… :20".

                  If you recopy one of the trouble files (the 1 out of 20), does it recopy with the same timestamp every time? If you delete the destination file (to avoid overwriting) first, and then re-copy: what timestamps does this get?
                  Yes, it's the same every time, with or without deleting the file between the tests.

                  One manual override is the Touch command, which can manually set the timestamp (after you verify the files are binary equal with a Compare Contents command).
                  Yes, Touch with option "copy from other side" works.

                  Btw., I noticed that for the above file, BC4 shows in the "Modified" column "12:19:21". The leading "12" is correct (time zone offset), but the 21 seconds is just the mid between the file's correct timestamp "12:19:22" on the source system and "12:19:20" after copied with BC4.

                  The same results can be observed with other affected files. The common pattern, for example:
                  The correct timestamp is … :16, BC4 shows … :15 and copies to the result of … :14.
                  Last edited by Carsten; 02-May-2018, 09:36 AM. Reason: Added result of testing with other files.

                    Comment

                    • Aaron
                      Team Scooter
                      • Oct 2007
                      • 15997
                      #6
                      Hello,

                      Ah, to clarify, the :21 second reporting in BC4: is this only on the downloaded file, or are both sides reporting as :21 (and hence, showing as equal in BC4's interface, despite differences when checked from the command line)?

                      And can you generate the timestamp in seconds on the source and destination after a BC4 transfer, using:
                      ls -l --time-style=+%s
                      and once more after using Filezilla (which, we assume will be exactly the same, but good to re-verify)?

                      While we don't have any timestamp specific troubleshooting in our log, it might be good to get a full connection log emailed to [email protected] (along with a link back to this forum thread), just to see if anything about the server/connection or transfer jump out as odd. A copy of a successful filezilla log might also help us compare and contrast. This would include initial connection and transfer of one trouble file.
                      Aaron P Scooter Software

                        Comment

                        • Zoë
                          Team Scooter
                          • Oct 2007
                          • 2666
                          #7
                          Can you verify that my understanding of the situation is correct?

                          Code:
                          Source date via stat                                    :22
Source date displayed in BC via SFTP                    :22
    
Dest date via stat after copy via BC                    :20
Dest date displayed in BC after copy                    :21
Dest date via stat after "touch/copy from other side"   :22
Dest date displayed in BC after "touch/copy..."         :22
                          • Is the issue only occuring for files that have a .0 ms value, and are there files with .0 ms that are working correctly?
                          • If you set one of the timestamps with the same value but with a ms value, does it start working correctly? You can do it in touch using something like `touch -d "2018-03-14 11:19:22.0001" filename`
                          • If you set one of the timestamps to the same value (with 0 ms) but plus or minus 1 and 2 seconds, does it still reproduce? (So, :20, :21, :22, :23, :24)


                          Are you able to run BC interactively on the Ubuntu 16 system, and if so:
                          • What does it display for the original file if you load it using a local path?
                          • If you have a correctly dated file on the Ubuntu 18 system and load a similar session where Ubuntu 16 is local and Ubuntu 18 is via SFTP, does the same issue occur?


                          SFTP does have a concept of optionally including subseconds or not, and I suspect that's the most likely place where something is going wrong, but we haven't been able to reproduce the issue in house yet.
                          Zoë P Scooter Software

                            Comment

                            • Carsten
                              Journeyman
                              • Oct 2015
                              • 18
                              #8
                              Hello Aaron,
                              hello Zoë,

                              ok, here is what I did for further testing. I also believe that this addresses all your questions:

                              1. Created test files on the Ubuntu 16.04 source system with `cp --preserve=timestamps …`, all but one from the problem file. The intention was to cover all the test cases that you mentioned above, explicitly using e.g. touch -d "2018-03-14 11:19:22.0" test-22.pdf to create the desired timestamps (except for test-fails.pdf and other-file-ok.pdf, which are the direct results of the cp --preserve=timestamps …). Result:

                              Code:
                              carsten@ProLiant-Ubuntu:~/LaufwerkD/test$ dir --full-time
insgesamt 16984
-rw-rw-r-- 1 carsten carsten  311926 2018-03-06 23:01:42.000000000 +0100 other-file-ok.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:20.000000000 +0100 test-20.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:21.000000000 +0100 test-21.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:22.000000000 +0100 test-22.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:23.000000000 +0100 test-23.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:24.000000000 +0100 test-24.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:25.000000000 +0100 test-25.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:26.000000000 +0100 test-26.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:27.000000000 +0100 test-27.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:22.000000000 +0100 test-fails.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:22.000100000 +0100 test-with-1ms.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:21.999999999 +0100 test-with-minus.pdf
                              2. Copied these files with BC4. See the attached screenshot. Here is the result on the target system:
                              Code:
                              carsten@red-glow:~/test$ dir --full-time
insgesamt 16984
-rw-rw-r-- 1 carsten carsten  311926 2018-03-06 23:01:42.000000000 +0100 other-file-ok.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:20.000000000 +0100 test-20.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:20.000000000 +0100 test-21.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:20.000000000 +0100 test-22.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:23.000000000 +0100 test-23.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:23.000000000 +0100 test-24.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:25.000000000 +0100 test-25.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:26.000000000 +0100 test-26.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:26.000000000 +0100 test-27.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:20.000000000 +0100 test-fails.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:20.000000000 +0100 test-with-1ms.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:20.000000000 +0100 test-with-minus.pdf
                              And the BC4 log:
                              Code:
                              03.05.2018 09:47:41  Username: carsten
03.05.2018 09:47:41  Connecting to vdzuggmrroo5k7e9.myfritz.net
03.05.2018 09:47:41  Server key [RSA 2048 9b:cc:b7:8a:2b:12:e0:2f:20:2b:f4:bf:93:6c:7e:2b]
03.05.2018 09:47:41  Public key authorization successful.
03.05.2018 09:47:42  Server software: OpenSSH_7.2p2 Ubuntu-4ubuntu2.4
03.05.2018 09:47:42  Compression: none
03.05.2018 09:47:42  Encryption: aes256-ctr
03.05.2018 09:47:42  MAC algorithm: hmac-sha2-512
03.05.2018 09:47:42  Connected to vdzuggmrroo5k7e9.myfritz.net port 22222
03.05.2018 09:47:42  Using SFTP version 3.
03.05.2018 09:47:42  Server information:
03.05.2018 09:47:42  Current Folder: /home/carsten
03.05.2018 09:47:42  Load comparison: /home/carsten/test <-> sftp://[email protected]:22222/LaufwerkD/test
03.05.2018 09:48:01  Connecting to vdzuggmrroo5k7e9.myfritz.net
03.05.2018 09:48:01  Connecting to vdzuggmrroo5k7e9.myfritz.net
03.05.2018 09:48:01  Connecting to vdzuggmrroo5k7e9.myfritz.net
03.05.2018 09:48:01  Server key [RSA 2048 9b:cc:b7:8a:2b:12:e0:2f:20:2b:f4:bf:93:6c:7e:2b]
03.05.2018 09:48:01  Server key [RSA 2048 9b:cc:b7:8a:2b:12:e0:2f:20:2b:f4:bf:93:6c:7e:2b]
03.05.2018 09:48:01  Server key [RSA 2048 9b:cc:b7:8a:2b:12:e0:2f:20:2b:f4:bf:93:6c:7e:2b]
03.05.2018 09:48:02  Public key authorization successful.
03.05.2018 09:48:02  Public key authorization successful.
03.05.2018 09:48:02  Public key authorization successful.
03.05.2018 09:48:03  Server software: OpenSSH_7.2p2 Ubuntu-4ubuntu2.4
03.05.2018 09:48:03  Compression: none
03.05.2018 09:48:03  Encryption: aes256-ctr
03.05.2018 09:48:03  MAC algorithm: hmac-sha2-512
03.05.2018 09:48:03  Connected to vdzuggmrroo5k7e9.myfritz.net port 22222
03.05.2018 09:48:03  Using SFTP version 3.
03.05.2018 09:48:03  Server information:
03.05.2018 09:48:03  Current Folder: /home/carsten
03.05.2018 09:48:03  Server software: OpenSSH_7.2p2 Ubuntu-4ubuntu2.4
03.05.2018 09:48:03  Compression: none
03.05.2018 09:48:03  Encryption: aes256-ctr
03.05.2018 09:48:03  MAC algorithm: hmac-sha2-512
03.05.2018 09:48:03  Connected to vdzuggmrroo5k7e9.myfritz.net port 22222
03.05.2018 09:48:03  Using SFTP version 3.
03.05.2018 09:48:03  Server information:
03.05.2018 09:48:03  Server software: OpenSSH_7.2p2 Ubuntu-4ubuntu2.4
03.05.2018 09:48:03  Compression: none
03.05.2018 09:48:03  Encryption: aes256-ctr
03.05.2018 09:48:03  MAC algorithm: hmac-sha2-512
03.05.2018 09:48:03  Connected to vdzuggmrroo5k7e9.myfritz.net port 22222
03.05.2018 09:48:03  Using SFTP version 3.
03.05.2018 09:48:03  Server information:
03.05.2018 09:48:03  Current Folder: /home/carsten
03.05.2018 09:48:03  Current Folder: /home/carsten
03.05.2018 09:51:48  Successfully copied 12 items.  Completed in 3 minutes, 47 seconds.
                              (Forum says that the text is too long. I'll continue in a separate post below.)
                              Attached Files
                              Last edited by Carsten; 03-May-2018, 03:55 AM. Reason: clarified the description

                                Comment

                                • Carsten
                                  Journeyman
                                  • Oct 2015
                                  • 18
                                  #9
                                  3. Copied the same files with FileZilla. Took a screenshot here as well (attached to the above post). Results:
                                  Code:
                                  carsten@red-glow:~/test-filezilla$ dir --full-time
insgesamt 16984
-rw-rw-r-- 1 carsten carsten  311926 2018-03-06 23:01:42.000000000 +0100 other-file-ok.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:20.000000000 +0100 test-20.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:21.000000000 +0100 test-21.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:22.000000000 +0100 test-22.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:23.000000000 +0100 test-23.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:24.000000000 +0100 test-24.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:25.000000000 +0100 test-25.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:26.000000000 +0100 test-26.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:27.000000000 +0100 test-27.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:22.000000000 +0100 test-fails.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:22.000000000 +0100 test-with-1ms.pdf
-rw-rw-r-- 1 carsten carsten 1549888 2018-03-14 11:19:21.000000000 +0100 test-with-minus.pdf
                                  FileZilla detailed log:
                                  Code:
                                  Status:	Verbinde mit vdzuggmrroo5k7e9.myfritz.net:22222...
Antwort:	fzSftp started, protocol_version=8
Befehl:	keyfile "/home/carsten/LaufwerkD/MyFiles/ssh-keys/carsten_id_rsa.ppk"
Befehl:	open "[email protected]" 22222
Status:	Connected to vdzuggmrroo5k7e9.myfritz.net
Status:	Empfange Verzeichnisinhalt...
Befehl:	pwd
Antwort:	Current directory is: "/home/carsten"
Befehl:	ls
Status:	Listing directory /home/carsten
Status:	Anzeigen des Verzeichnisinhalts für "/home/carsten" abgeschlossen
Status:	Empfange Verzeichnisinhalt für "/home/carsten/LaufwerkD"...
Befehl:	cd "/home/carsten/LaufwerkD"
Antwort:	New directory is: "/home/carsten/LaufwerkD"
Befehl:	ls
Status:	Listing directory /home/carsten/LaufwerkD
Status:	Anzeigen des Verzeichnisinhalts für "/home/carsten/LaufwerkD" abgeschlossen
Status:	Empfange Verzeichnisinhalt für "/home/carsten/LaufwerkD/test"...
Befehl:	cd "/home/carsten/LaufwerkD/test"
Antwort:	New directory is: "/home/carsten/LaufwerkD/test"
Befehl:	ls
Status:	Listing directory /home/carsten/LaufwerkD/test
Status:	Anzeigen des Verzeichnisinhalts für "/home/carsten/LaufwerkD/test" abgeschlossen
Status:	Verbinde mit vdzuggmrroo5k7e9.myfritz.net:22222...
Antwort:	fzSftp started, protocol_version=8
Befehl:	keyfile "/home/carsten/LaufwerkD/MyFiles/ssh-keys/carsten_id_rsa.ppk"
Befehl:	open "[email protected]" 22222
Status:	Verbinde mit vdzuggmrroo5k7e9.myfritz.net:22222...
Antwort:	fzSftp started, protocol_version=8
Befehl:	keyfile "/home/carsten/LaufwerkD/MyFiles/ssh-keys/carsten_id_rsa.ppk"
Befehl:	open "[email protected]" 22222
Status:	Connected to vdzuggmrroo5k7e9.myfritz.net
Status:	Connected to vdzuggmrroo5k7e9.myfritz.net
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-20.pdf
Befehl:	cd "/home/carsten/LaufwerkD/test"
Status:	Starte Download von /home/carsten/LaufwerkD/test/other-file-ok.pdf
Befehl:	cd "/home/carsten/LaufwerkD/test"
Antwort:	New directory is: "/home/carsten/LaufwerkD/test"
Befehl:	get "test-20.pdf" "/home/carsten/test-filezilla/test-20.pdf"
Antwort:	New directory is: "/home/carsten/LaufwerkD/test"
Befehl:	get "other-file-ok.pdf" "/home/carsten/test-filezilla/other-file-ok.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-20.pdf => local:/home/carsten/test-filezilla/test-20.pdf
Befehl:	remote:/home/carsten/LaufwerkD/test/other-file-ok.pdf => local:/home/carsten/test-filezilla/other-file-ok.pdf
Status:	Dateiübertragung erfolgreich, 312,0 KB in 9 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-21.pdf
Befehl:	get "test-21.pdf" "/home/carsten/test-filezilla/test-21.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-21.pdf => local:/home/carsten/test-filezilla/test-21.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 35 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-22.pdf
Befehl:	get "test-22.pdf" "/home/carsten/test-filezilla/test-22.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-22.pdf => local:/home/carsten/test-filezilla/test-22.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 42 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-23.pdf
Befehl:	get "test-23.pdf" "/home/carsten/test-filezilla/test-23.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-23.pdf => local:/home/carsten/test-filezilla/test-23.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 44 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-24.pdf
Befehl:	get "test-24.pdf" "/home/carsten/test-filezilla/test-24.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-24.pdf => local:/home/carsten/test-filezilla/test-24.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 36 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-25.pdf
Befehl:	get "test-25.pdf" "/home/carsten/test-filezilla/test-25.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-25.pdf => local:/home/carsten/test-filezilla/test-25.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 34 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-26.pdf
Befehl:	get "test-26.pdf" "/home/carsten/test-filezilla/test-26.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-26.pdf => local:/home/carsten/test-filezilla/test-26.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 43 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-27.pdf
Befehl:	get "test-27.pdf" "/home/carsten/test-filezilla/test-27.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-27.pdf => local:/home/carsten/test-filezilla/test-27.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 42 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-fails.pdf
Befehl:	get "test-fails.pdf" "/home/carsten/test-filezilla/test-fails.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-fails.pdf => local:/home/carsten/test-filezilla/test-fails.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 38 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-with-1ms.pdf
Befehl:	get "test-with-1ms.pdf" "/home/carsten/test-filezilla/test-with-1ms.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-with-1ms.pdf => local:/home/carsten/test-filezilla/test-with-1ms.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 38 Sekunden übertragen
Status:	Starte Download von /home/carsten/LaufwerkD/test/test-with-minus.pdf
Befehl:	get "test-with-minus.pdf" "/home/carsten/test-filezilla/test-with-minus.pdf"
Befehl:	remote:/home/carsten/LaufwerkD/test/test-with-minus.pdf => local:/home/carsten/test-filezilla/test-with-minus.pdf
Status:	Dateiübertragung erfolgreich, 1,6 MB in 46 Sekunden übertragen
Status:	Dateiübertragung erfolgreich, 1,6 MB in 29 Sekunden übertragen
Status:	Verbindung zum Server getrennt
Status:	Verbindung zum Server getrennt
                                  Well, the timestamps in BC4 look really weird to me…
                                  Last edited by Carsten; 03-May-2018, 04:01 AM. Reason: Fixed typo in prose

                                    Comment

                                    • Zoë
                                      Team Scooter
                                      • Oct 2007
                                      • 2666
                                      #10
                                      Thank you for all of the testing. I tracked the issue down to a lossy conversion in our timestamp handling code. It will be fixed in the next release (v4.2.5).

                                      There will still be two limitations you might notice: (1) BC only works with timestamps down to the second, so any millisecond values may be truncated when copying files, and (2) timezone conversions use whatever the current timezone adjustment is, rather than whatever was active on that date. That actually shows up in your test case, since 3/6 and 3/14 are on opposite sides of the DST switch. That mostly only affects the display, but it's something to be aware of if you compare 'ls' and BC listings. Those are both things we want to fix, but they'll touch enough code that we won't be able to release them as part a v4 patch.
                                      Zoë P Scooter Software

                                        Comment

                                        • Carsten
                                          Journeyman
                                          • Oct 2015
                                          • 18
                                          #11
                                          Thank you!

                                          Is there any release date set for 4.2.5 yet?

                                            Comment

                                            • Aaron
                                              Team Scooter
                                              • Oct 2007
                                              • 15997
                                              #12
                                              Hello,

                                              We do not have a set date for 4.2.5, but we working towards a candidate, which will then have to pass QA.
                                              Aaron P Scooter Software

                                                Comment

                                                • Chris
                                                  Team Scooter
                                                  • Oct 2007
                                                  • 5538
                                                  #13
                                                  BC 4.2.5 is now available, it has a fix for the timestamp issue you reported.
                                                  Chris K Scooter Software

                                                    Comment

                                                    • Carsten
                                                      Journeyman
                                                      • Oct 2015
                                                      • 18
                                                      #14
                                                      Thanks for your efforts and for updating the thread!

                                                      However, I've just tested with BC Version 4.2.5 (build 23088), and I see quasi the same results as before:
                                                      copying a set of files yields wrong timestamps on the target.

                                                      This is a newly made test case, but exactly alike the former testcase:

                                                      Remote / source system:
                                                      Code:
                                                      carsten@ProLiant-Ubuntu:~/LaufwerkD/test$ dir --full-time
insgesamt 132
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:20.000000000 +0100 test-20.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:21.000000000 +0100 test-21-0000.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:21.999999999 +0100 test-21-9999.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:22.000000000 +0100 test-22-0000.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:22.000100000 +0100 test-22-0001.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:23.000000000 +0100 test-23.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:24.000000000 +0100 test-24.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:25.000000000 +0100 test-25.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:26.000000000 +0100 test-26.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:27.000000000 +0100 test-27.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-06-20 15:36:24.606735287 +0200 test.odg
                                                      Local / target system to which the above files were copied with BC 4.2.5:
                                                      Code:
                                                      carsten@red-glow:~/test-bc-4.2.5$ dir --full-time
insgesamt 132
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:20.000000000 +0100 test-20.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:20.000000000 +0100 test-21-0000.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:20.000000000 +0100 test-21-9999.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:21.000000000 +0100 test-22-0000.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:21.000000000 +0100 test-22-0001.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:23.000000000 +0100 test-23.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:23.000000000 +0100 test-24.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:25.000000000 +0100 test-25.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:26.000000000 +0100 test-26.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-03-14 11:19:26.000000000 +0100 test-27.odg
-rw-rw-r-- 1 carsten carsten 10203 2018-06-20 15:36:23.000000000 +0200 test.odg
                                                      Can you please have a look into this?

                                                        Comment

                                                        • Aaron
                                                          Team Scooter
                                                          • Oct 2007
                                                          • 15997
                                                          #15
                                                          Thanks. We have some fixes implemented, which help when loading/comparing, but transferring seems to still introduce some issues. We'll re-open and investigate.
                                                          Aaron P Scooter Software

                                                            Comment

                                                            Previous 1 2 template Next
                                                            Working...