--inplace can prevent rsync from running you out of disk space when updating a large file. But you don't want to mix it with --update. Judging by your other message you think that --update means modiify files that are already on both ends. It doesn't. It only means that rsync is forbidden to change files on the target that have a newer timestamp than is on the source. If you don't need that then you don't want --update.
Also, when in doubt, --dry-run. On 9/11/21 11:51 PM, hancooper wrote: > ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ > On Sunday, September 12, 2021 2:34 AM, hancooper via rsync > <rsync@lists.samba.org> wrote: > >> ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ >> On Sunday, September 12, 2021 2:03 AM, Kevin Korb k...@sanitarium.net wrote: >> >>> I thought I did elaborate. If it is a problem for you then maybe you >>> shouldn't be using --update. Or you should let rsync delete incomplete >>> files upon abort as it does by default. >> >> I am using the following >> >> rsync -av --inplace --update --log-file=/media/hagbard/hc/snapshot/test1.log >> /media/hagbard/hc1/snapshot/test1 /media/hagbard/hc/snapshot/ >> >> I want that if I run the rsync command again, the files already transferred >> are not resent, thus I used --update. > > Can I re-run without the --update option? Am I right to say that the files > already successfully transferred will not be re-sent again using -av > --inplace ? > > What can I do for the files which have been sent but got rsync aborted during > transfer? > > >>> On 9/11/21 9:29 PM, hancooper wrote: >>> >>>> ‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐ >>>> On Saturday, September 11, 2021 11:20 PM, Kevin Korb via rsync >>>> rsync@lists.samba.org wrote: >>>> >>>>> --archive is all you really need. I actually wish --archive was the >>>>> default because it is all most people need and with the exception of >>>>> writing to a FAT filesystem it is almost always needed. >>>>> --append is for very special cases and should only be used if you really >>>>> know you need it and why. --append-verify only exists because people >>>>> seem to think they need --append and get annoyed that it corrupts their >>>>> files. >>>>> --update is sometimes helpful however it interacts badly with --partial >>>>> and --inplace. >>>> >>>> That's right. Can you elaborate how --update interacts badly with --partial >>>> and --inplace? >>>> >>>>> Since rsync doesn't set the timestamp of a file until it >>>>> is finished then any file left incomplete by an aborted rsync will have >>>>> the timestamp of the abort not the source file. Therefore it will be >>>>> newer than the source file and unless the source is updated rsync >>>>> --update will never complete the file. >>>> >>>> Is there a solution to this problem and should we consider it a bug? >>>> I am encountering this, and it's a real problem for me if files are >>>> not completed. >>>> >>>>> On 9/11/21 6:50 PM, hancooper via rsync wrote: >>>>> >>>>>> I am struggling to understand exactly what the rsync options --update >>>>>> and --append-verify do. >>>>>> Doing info rsync gives >>>>>> -u, --update >>>>>> This forces rsync to skip any files which exist on the destina‐ >>>>>> tion and have a modified time that is newer than the source >>>>>> file. (If an existing destination file has a modification time >>>>>> equal to the source file’s, it will be updated if the sizes are >>>>>> different.) >>>>>> --append-verify >>>>>> This works just like the --append option, but the existing data >>>>>> on the receiving side is included in the full-file checksum >>>>>> verification step, which will cause a file to be resent if the >>>>>> final verification step fails (rsync uses a normal, non-append‐ >>>>>> ing --inplace transfer for the resend). >>>>>> I am using rsync to transfer directories recursively. There are times >>>>>> where I have to stop the rsync transfer, and resume the transfer a few >>>>>> hours or days later, without affecting the already transferred files at >>>>>> destination. >>>>>> I also got some files that return errors by rsync, such as >>>>>> rsync: read errors mapping >>>>>> "/media/hc1/a1-chaos/amvib/IACL-2017-07-19T00:00:00-2017-07-19T23:59:59.mseed": >>>>>> Input/output error (5) >>>>>> I would like to retry the transfer of these files, at a later time too, >>>>>> without affecting the files that had been transferred successfully. >>> >>> -- >>> ~-,..,-~'`^`'~-,..,-~'`^`'~-,..,-~'`^`'~-,..,-~'`^`'~-,.., >>> >>> Kevin Korb Phone: (407) 252-6853 >>> >>> Systems Administrator Internet: >>> >>> FutureQuest, Inc. ke...@futurequest.net (work) >>> >>> Orlando, Florida k...@sanitarium.net (personal) >>> >>> Web page: https://sanitarium.net/ >>> >>> PGP public key available on web site. >>> >>> ~-,..,-~'`^`'~-,..,-~'`^`'~-,..,-~'`^`'~-,..,-~'`^`'~-,.., >> >> -- >> >> Please use reply-all for most replies to avoid omitting the mailing list. >> To unsubscribe or change options: >> https://lists.samba.org/mailman/listinfo/rsync >> Before posting, read: http://www.catb.org/~esr/faqs/smart-questions.html > -- ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._., Kevin Korb Phone: (407) 252-6853 Systems Administrator Internet: FutureQuest, Inc. ke...@futurequest.net (work) Orlando, Florida k...@sanitarium.net (personal) Web page: https://sanitarium.net/ PGP public key available on web site. ~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._., -- Please use reply-all for most replies to avoid omitting the mailing list. To unsubscribe or change options: https://lists.samba.org/mailman/listinfo/rsync Before posting, read: http://www.catb.org/~esr/faqs/smart-questions.html