FLAM® Issue Tracker

View Issue Details Jump to Notes ] Issue History ] Print ]
IDProjectCategoryView StatusDate SubmittedLast Update
0000793FL52.2 Subprogram FLUC (CONV)public2016-03-02 13:342018-06-18 09:46
ReporterFalk Reichbott 
Assigned ToFalk Reichbott 
PrioritynormalSeverityfeatureReproducibilityN/A
StatusassignedResolutionopen 
PlatformGeneralOSGeneralOS VersionGeneral
Product Version5.1.10 
Target Version5.1.24Fixed in Version 
Summary0000793: Support receipt and restart cabebility as part of the SSH support
DescriptionFor remote read and write of date a receipt for complete and a restart in case of incomplete transfer must be possible.

For restart a parameter point to a file name where the XCNV restart command is written. The restart command contains the name of the temproary file containing the rest of the data, which was not written to the other node. The restart command reads this file binary, append this to the incomplete target file on the other node and deletes the temporary file. This restart command generates again a new restart command if this transfer is not complete aso.

After each transfer a receipt will be written to the log and a special return code will be set if a restart is required.

The restart can not be based on the original file, because a lot of conversios can not be reproduced to append the missing rest to the file on the other node.

Means the conversion must be done and the rest of the file which was not transfered must be written to a temproary file to restart the transfer.

In a loop based on the special return code the restart cabebility can be used in scripts to automate this process.

Example:

flcl conv read.file='DD:INPUT' write.text(method=UNIX CCSID=UTF-8 comp.gzip() encr.pgp() file=f'url.txt' resend='resend.txt')

while ($?==SC) {
   flcl xcnv=resend.txt
}

if ($?!=0) {
   incomplete fransfer
}
TagsNo tags attached.
Attached Files

- Relationships

-  Notes
(0001013)
Falk Reichbott (administrator)
2016-07-26 15:32

Eine impliziete Lösung mit einem Reconnect-Counter (default ist 0) ist womöglich einfacher. Ist der Reconnect-Counter größer 0, dann werden alle Daten zusätzlich in eine temporäre Datei geschrieben. Wenn die connection wegbricht bzw. fstat remote eine kleiner Filesize zurückgibt, wird Reconnect-Count oft versucht den fehlende Rest automatisch noch zu schreiben. Erst wenn dies nicht gelinkt, wird der Name der temporären Datei zusammen mit einem Fehler und dem Resend-Kommando ausgegeben.

- Issue History
Date Modified Username Field Change
2016-03-02 13:34 Falk Reichbott New Issue
2016-03-02 13:34 Falk Reichbott Status new => assigned
2016-03-02 13:34 Falk Reichbott Assigned To => Falk Reichbott
2016-07-26 15:32 Falk Reichbott Note Added: 0001013
2016-08-18 10:51 Falk Reichbott Target Version 5.1.13 => 5.1.14
2016-11-16 09:01 Falk Reichbott Target Version 5.1.14 => 5.1.15
2017-02-27 14:21 Falk Reichbott Target Version 5.1.15 => 5.1.16
2017-05-10 13:39 Falk Reichbott Target Version 5.1.16 => 5.1.18
2018-06-18 09:46 Falk Reichbott Target Version 5.1.18 => 5.1.24


Copyright © 2000 - 2020 MantisBT Team
Powered by Mantis Bugtracker