RE: Force "ls" to show .exe extension
Paul sent the following at Tuesday, January 06, 2015 7:12 PM >I'm wading through many files in two file trees. In particular, I'm >looking at corresponding directories in the two trees where "diff -qr" >revealed differences. I want the absolute truth of what the filename is >with minimal distractions about how to achieve that. Then, I can focus >on figuring how those files came about, and how the differences arose. Not a Cygwin solution but the following should give real names. cmd /c dir /b /a: (The /a: makes sure that hidden files are listed.) - Barry - Disclaimer: Statements made herein are not made on behalf of NIAID. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: WinXP Pro Cygwin Digest-MD4-1.9 /bin/sh: gcc-4: command not found
On 01/05/2015 08:13 PM, Larry Hall (Cygwin) wrote: gcc-4 is an old moniker for the current gcc, back when Cygwin had both 3 and 4 versions of gcc. This is no longer true and hasn't been for quite some time. Your options at this point are: 1. Create a link to gcc and call it gcc-4. 2. Try the experimental 5.18 version. 3. Install the x86_64 version of Cygwin (requires x86_64 version of Windows). Okay -- that was the correct idea (I also needed to install 'g++' and create a link 'g++-4'). Thanks! David 2015-01-06 17:39:22 Administrator@dc8ct591 ~ $ which gcc /usr/bin/gcc 2015-01-06 17:39:22 Administrator@dc8ct591 ~ $ ll /usr/bin/gcc -rwxr-xr-x 3 Administrator None 696871 2014/11/11 12:19:22 /usr/bin/gcc* 2015-01-06 17:39:38 Administrator@dc8ct591 ~ $ ln /usr/bin/gcc /usr/bin/gcc-4 2015-01-06 17:39:57 Administrator@dc8ct591 ~ $ which gcc-4 /usr/bin/gcc-4 2015-01-06 17:40:00 Administrator@dc8ct591 ~ $ cpan Digest::MD4 CPAN: Storable loaded ok (v2.27) Going to read '/home/Administrator/.cpan/Metadata' Database was generated on Mon, 24 Nov 2014 11:53:19 GMT CPAN: LWP::UserAgent loaded ok (v6.04) Fetching with LWP: http://ftp.wayne.edu/CPAN/authors/01mailrc.txt.gz CPAN: YAML loaded ok (v0.80) Going to read '/home/Administrator/.cpan/sources/authors/01mailrc.txt.gz' CPAN: Compress::Zlib loaded ok (v2.033) DONE Fetching with LWP: http://ftp.wayne.edu/CPAN/modules/02packages.details.txt.gz Going to read '/home/Administrator/.cpan/sources/modules/02packages.details.txt.gz' Database was generated on Tue, 06 Jan 2015 12:29:02 GMT .. New CPAN.pm version (v2.05) available. [Currently running version is v1.960001] You might want to try install CPAN reload cpan to both upgrade CPAN.pm and run the new version without leaving the current session. ..DONE Fetching with LWP: http://ftp.wayne.edu/CPAN/modules/03modlist.data.gz Going to read '/home/Administrator/.cpan/sources/modules/03modlist.data.gz' DONE Going to write /home/Administrator/.cpan/Metadata Running install for module 'Digest::MD4' Running make for M/MI/MIKEM/DigestMD4/Digest-MD4-1.9.tar.gz CPAN: Digest::SHA loaded ok (v5.71) Checksum for /home/Administrator/.cpan/sources/authors/id/M/MI/MIKEM/DigestMD4/Digest-MD4-1.9.tar.gz ok CPAN: File::Temp loaded ok (v0.22) CPAN: Parse::CPAN::Meta loaded ok (v1.4405) CPAN: CPAN::Meta loaded ok (v2.120351) CPAN: Module::CoreList loaded ok (v2.49_02) CPAN.pm: Going to build M/MI/MIKEM/DigestMD4/Digest-MD4-1.9.tar.gz Checking if your kit is complete... Looks good Generating a Unix-style Makefile Writing Makefile for Digest::MD4 Writing MYMETA.yml and MYMETA.json cp MD4.pm blib/lib/Digest/MD4.pm Running Mkbootstrap for Digest::MD4 () chmod 644 "MD4.bs" "/usr/bin/perl.exe" "/usr/lib/perl5/5.14/ExtUtils/xsubpp" -typemap "/usr/lib/perl5/5.14/ExtUtils/typemap" -typemap "typemap" MD4.xs > MD4.xsc && mv MD4.xsc MD4.c gcc-4 -c -DPERL_USE_SAFE_PUTENV -U__STRICT_ANSI__ -g -fno-strict-aliasing -pipe -fstack-protector -DUSEIMPORTLIB -O3 -DVERSION=\"1.9\" -DXS_VERSION=\"1.9\" "-I/usr/lib/perl5/5.14/i686-cygwin-threads-64int/CORE" MD4.c rm -f blib/arch/auto/Digest/MD4/MD4.dll g++-4 --shared -Wl,--enable-auto-import -Wl,--export-all-symbols -Wl,--enable-auto-image-base -L/usr/local/lib -fstack-protector MD4.o -o blib/arch/auto/Digest/MD4/MD4.dll \ /usr/lib/perl5/5.14/i686-cygwin-threads-64int/CORE/cygperl5_14.dll\ make: g++-4: Command not found Makefile:464: recipe for target 'blib/arch/auto/Digest/MD4/MD4.dll' failed make: *** [blib/arch/auto/Digest/MD4/MD4.dll] Error 127 MIKEM/DigestMD4/Digest-MD4-1.9.tar.gz /usr/bin/make -- NOT OK Running make test Can't test without successful make Running make install Make had returned bad status, install seems impossible 2015-01-06 17:41:36 Administrator@dc8ct591 ~ $ which g++ which: no g++ in (/home/Administrator/perl5/bin:/home/Administrator/bin:/home/dpchrist/jdk1.7.0_25/bin:/usr/local/bin:/usr/bin:/bin:/cygdrive/c/WINDOWS/system32:/cygdrive/c/WINDOWS:/cygdrive/c/WINDOWS/System32/Wbem:/cygdrive/c/Program Files/Intel/DMIX:/cygdrive/c/Program Files/Common Files/Acronis/SnapAPI:/usr/sbin) 2015-01-06 17:41:46 Administrator@dc8ct591 ~ $ # install gcc-g++ 2015-01-06 17:43:22 Administrator@dc8ct591 ~ $ which g++ /usr/bin/g++ 2015-01-06 17:44:56 Administrator@dc8ct591 ~ $ ln /usr/bin/g++ /usr/bin/g++-4 2015-01-06 17:45:11 Administrator@dc8ct591 ~ $ cpan Digest::MD4 CPAN: Storable loaded ok (v2.27) Going to read '/home/Administrator/.cpan/Metadata' Database was generated on Tue, 06 Jan 2015 12:29:02 GMT Running install for module 'Digest::MD4' Running make for M/MI/MIKEM/DigestMD4/Digest-MD4-1.9.tar.gz CPAN: Digest::SHA loaded ok (v5.71) CPAN: Compress::Zlib loaded ok (v2.033) Checksum for /home/Administrator/.cpan/sources/authors/id/M/MI/MIKEM/DigestMD4/
Re: Force "ls" to show .exe extension
Greetings, Paul! >>> ...if I have ~/bin/pdfcrop.exe, the command "ls ~/bin/pdfcrop" >>> shows pdfcrop rather than pdfcrop.exe. Is there any way to force >>> ls to show the full filename (including extension) if it matched >>> the ls argument, even if the ls argument doesn't specify the >>> extension? I read >>> http://cygwin.com/cygwin-ug-net/using-specialnames.html, which >>> helps explain the situation, but not a solution. >> >> I see people are diving deep to find a "solution", but noone has >> asked, what is the problem they are trying to solve here. So, let >> me ask it: what is the problem? > Hello, Andrey, > I'm wading through many files in two file trees. In particular, I'm > looking at corresponding directories in the two trees where "diff -qr" > revealed differences. I want the absolute truth of what the filename > is with minimal distrations about how to achieve that. Then, I can > focus on figuring how those files came about, and how the differences > arose. Use more appropriate tools. I.e. Far manager have standard file/directory comparison plugin. Which can quickly compare paths and or contents of the files. However, it will not show differences inside files, only show that they are different, but you can always augment it with view: Sorry for my terrible english... -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Force "ls" to show .exe extension
Andrey Repin yandex.ru> writes: >> ...if I have ~/bin/pdfcrop.exe, the command "ls ~/bin/pdfcrop" >> shows pdfcrop rather than pdfcrop.exe. Is there any way to force >> ls to show the full filename (including extension) if it matched >> the ls argument, even if the ls argument doesn't specify the >> extension? I read >> http://cygwin.com/cygwin-ug-net/using-specialnames.html, which >> helps explain the situation, but not a solution. > > I see people are diving deep to find a "solution", but noone has > asked, what is the problem they are trying to solve here. So, let > me ask it: what is the problem? Hello, Andrey, I'm wading through many files in two file trees. In particular, I'm looking at corresponding directories in the two trees where "diff -qr" revealed differences. I want the absolute truth of what the filename is with minimal distrations about how to achieve that. Then, I can focus on figuring how those files came about, and how the differences arose. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Force
On 01/06/2015 05:06 PM, Paul wrote: > | alias xargs='xargs ' > | > | Then the alias expansion of xargs will in turn allow alias expansion > | of the next argument. (Except that you then have to also create > | trailing-space aliases for all options you commonly pass to xargs > | between 'xargs' and the final command name). > > I'm not sure what you mean by needing trailing space aliases for > common xargs options, but I'm going to take that as a warning of > dragons lurking in that direction and avoid it. I mean that: alias xargs='xargs ' alias ls='ls --append-exe' find -pa pdfcrop | xargs ls will execute 'ls --append-exe', but alias xargs='xargs ' alias ls='ls --append-exe' find -pa pdfcrop -print0 | xargs -0 ls will not, unless you also: alias -- -0='-0 ' -- Eric Blake eblake redhat com+1-919-301-3266 Libvirt virtualization library http://libvirt.org signature.asc Description: OpenPGP digital signature
Re: Force
Eric Blake redhat.com> writes: |On 01/06/2015 02:28 PM, Paul wrote: |>Paul gmail.com> writes: |>> Both solutions are great. I'll set the --append-exe in my bash |>> aliases, and for systems outside of my normal working environment |>> (e.g., working with someone on their unix sessions), I know I can |>> force display of .exe using asterisk. |> |> Drat. If I pipe files to 'xargs ls', the unaliased ls command is |> used: |> |>type -pa pdfcrop | xargs ls | | alias xargs='xargs ' | | Then the alias expansion of xargs will in turn allow alias expansion | of the next argument. (Except that you then have to also create | trailing-space aliases for all options you commonly pass to xargs | between 'xargs' and the final command name). I'm not sure what you mean by needing trailing space aliases for common xargs options, but I'm going to take that as a warning of dragons lurking in that direction and avoid it. | Sadly, xargs is one of the cases where shell functions won't help | (xargs doesn't execute the shell function). Your other solution is | to modify $PATH to point to a directory under your control as the | first thing, where 'cat /your/ls' contains: | | #!/bin/sh | exec /bin/ls --append-exe "$ " | | such that your script then gets picked up by xargs, and you no | longer have to worry about aliases. OK, I'll keep that one in mind -- wrapper scripts rather than aliases and functions. Thanks. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Cygwin script wrapper (native)
Greetings, All! > I think I've finally ironed out most of the issues I had with it. (Infinite > number of arguments, quoting, etc.) > The only known issues that cause script to break seems to be the same as you > would expect with direct script executions - uneven/misplaced quotation marks. Except the line 31, which should read FOR /F "usebackq delims=" %%a IN (`cygpath.exe -u "%~1"`) DO ( else the paths containing spaces are split in multiple tokens. -- WBR, Andrey Repin (anrdae...@yandex.ru) 07.01.2015, <02:16> Sorry for my terrible english... -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Force "ls" to show .exe extension
Greetings, Paul! > Right now, if I have ~/bin/pdfcrop.exe, the command "ls ~/bin/pdfcrop" > shows pdfcrop rather than pdfcrop.exe. Is there any way to force ls > to show the full filename (including extension) if it matched the ls > argument, even if the ls argument doesn't specify the extension? I > read http://cygwin.com/cygwin-ug-net/using-specialnames.html, which > helps explain the situation, but not a solution. I see people are diving deep to find a "solution", but noone has asked, what is the problem they are trying to solve here. So, let me ask it: what is the problem? -- WBR, Andrey Repin (anrdae...@yandex.ru) 07.01.2015, <01:31> Sorry for my terrible english... -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Force "ls" to show .exe extension
On 01/06/2015 02:28 PM, Paul wrote: > Paul gmail.com> writes: >> Both solutions are great. I'll set the --append-exe in my bash >> aliases, and for systems outside of my normal working environment >> (e.g., working with someone on their unix sessions), I know I can >> force display of .exe using asterisk. > > Drat. If I pipe files to 'xargs ls', the unaliased ls command is used: > >type -pa pdfcrop | xargs ls alias xargs='xargs ' Then the alias expansion of xargs will in turn allow alias expansion of the next argument. (Except that you then have to also create trailing-space aliases for all options you commonly pass to xargs between 'xargs' and the final command name). Sadly, xargs is one of the cases where shell functions won't help (xargs doesn't execute the shell function). Your other solution is to modify $PATH to point to a directory under your control as the first thing, where 'cat /your/ls' contains: #!/bin/sh exec /bin/ls --append-exe "$@" such that your script then gets picked up by xargs, and you no longer have to worry about aliases. -- Eric Blake eblake redhat com+1-919-301-3266 Libvirt virtualization library http://libvirt.org signature.asc Description: OpenPGP digital signature
Re: [RSYNC] bad modify/change time set by rsync ?
Greetings, LEGOND Fabrice! >> --modify-window >> >> When comparing two timestamps, rsync treats the timestamps as being >> equal if they differ by no more than the modify-window value. This >> is normally 0 (for an exact match), but you may find it useful to >> set this to a larger value in some situations. In particular, when >> transferring to or from an MS Windows FAT filesystem (which >> represents times with a 2-second resolution), --modify-window=1 is >> useful (allowing times to differ by up to 1 second). >> >> Yes, (ex)FAT just doesn't have the resolution to represent some >> timestamps, specifically it can only record even seconds. > Yes, that were also a guess at one moment in my testing time, but was > discarded because, as I said, I had the very same problem on a ntfs > partitions which have far greater precision. So is this normal ? This is normal when your timezone is incorrect on one of the hosts. Means, conversion to UTC result in different timestamps. P.S. Please don't break threads. -- WBR, Andrey Repin (anrdae...@yandex.ru) 07.01.2015, <00:15> Sorry for my terrible english... -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Force "ls" to show .exe extension
Paul gmail.com> writes: > Both solutions are great. I'll set the --append-exe in my bash > aliases, and for systems outside of my normal working environment > (e.g., working with someone on their unix sessions), I know I can > force display of .exe using asterisk. Drat. If I pipe files to 'xargs ls', the unaliased ls command is used: type -pa pdfcrop | xargs ls I can always append asterisks to each pdfcrop path using sed, or even explicitly type --append-exe [:(] . -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Force "ls" to show .exe extension
Tom Robinson gmail.com> writes: >If you don't want to specify the extension, can you specify as >asterisk? > >[3236 CBGSAS04:~/Documents]$ touch name.exe > >[3237 CBGSAS04:~/Documents]$ ls -l name >-rw-r--r--+ 1 cbg.tom Domain Users 0 Jan 7 09:34 name > >[3238 CBGSAS04:~/Documents]$ ls -l name.exe >-rw-r--r--+ 1 cbg.tom Domain Users 0 Jan 7 09:34 name.exe > >[3239 CBGSAS04:~/Documents]$ ls -l name* >-rw-r--r--+ 1 cbg.tom Domain Users 0 Jan 7 09:34 name.exe Yaakov Selkowitz cygwin.com> writes: > ls --append-exe ~/bin/pdfcrop Tom, Yaakov, Both solutions are great. I'll set the --append-exe in my bash aliases, and for systems outside of my normal working environment (e.g., working with someone on their unix sessions), I know I can force display of .exe using asterisk. Thanks! -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Force "ls" to show .exe extension
On 2015-01-06 13:34, Paul wrote: Right now, if I have ~/bin/pdfcrop.exe, the command "ls ~/bin/pdfcrop" shows pdfcrop rather than pdfcrop.exe. Is there any way to force ls to show the full filename (including extension) if it matched the ls argument, even if the ls argument doesn't specify the extension? I read http://cygwin.com/cygwin-ug-net/using-specialnames.html, which helps explain the situation, but not a solution. ls --append-exe ~/bin/pdfcrop -- Yaakov -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Force "ls" to show .exe extension
If you don't want to specify the extension, can you specify as asterisk? [3236 CBGSAS04:~/Documents]$ touch name.exe [3237 CBGSAS04:~/Documents]$ ls -l name -rw-r--r--+ 1 cbg.tom Domain Users 0 Jan 7 09:34 name [3238 CBGSAS04:~/Documents]$ ls -l name.exe -rw-r--r--+ 1 cbg.tom Domain Users 0 Jan 7 09:34 name.exe [3239 CBGSAS04:~/Documents]$ ls -l name* -rw-r--r--+ 1 cbg.tom Domain Users 0 Jan 7 09:34 name.exe > On 2015-01-07, at 08:34, Paul wrote: > > Right now, if I have ~/bin/pdfcrop.exe, the command "ls ~/bin/pdfcrop" > shows pdfcrop rather than pdfcrop.exe. Is there any way to force ls > to show the full filename (including extension) if it matched the ls > argument, even if the ls argument doesn't specify the extension? I > read http://cygwin.com/cygwin-ug-net/using-specialnames.html, which > helps explain the situation, but not a solution. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
RE: SSHd configuration problems (System error 1376) - CONFIRMED
Ok, I'm hopeful this message is going to reach the correct thread. Thanks for your patience, I was accessing and posting to the discussion via web browser which is how I discovered it. Now that I'm properly subscribed to the list and using my email client... on to business. Regarding sshd configuration using the ssh-host-config command resulting in system error 1376: >System error 1376 has occurred. >The specified local group does not exist. >Adding user 'cyg_server' to local group 'root' failed! This result is fairly new. Success was the normal result as recently as a fresh Cygwin64 install on November 7. In the error scenario, if the user proceeds with the configuration the sshd service is assigned to the local windows SYSTEM user (per default choice provided if I remember correctly). Tied to the SYSTEM user account, the sshd service will start and run, but when remote login is attempted, the ssh connection is terminated immediately by host. As mentioned, I have both a "root" and "Administrators" entries in my etc\group file as the result of the default install procedure. There is no Windows group 'root'. The group file was not tampered with by hand. I see the same 'root' entry in the group file of the server we successfully built on Nov 7. The fact of the root entry in the group file may not have anything to do with the 1376 problem except that the local group 'root' is mentioned in the error message. In the error scenario, the Windows user account cyg_server is created. At conclusion of the process it is a member of the Windows "Users" group only. I attempted this work-around: 1) Manually making the Windows cyg_server account a member of the Windows Administrators group (via Windows User management interface) 2) Then re-running mkpasswd and confirming that user cyg_server was associated with the Administrators group 3) Then assigning the sshd service to start using the .\cyg_server user (via Windows interface) instead of SYSTEM. Shutting down and then restarting the service. The result was the service attempting to start, and then shutting itself down immediately. This is the work-around as I understand it from the posts above but perhaps I am missing a key step. Anyway, I'd like to help with solving the underlying problem if possible. Attached is the result of: cygcheck -s -v -r > cygcheck.out on my server with sensitive user and site-specific information redacted. You'll find "XX" anywhere I made a change to mask that info. Thanks for your time and any of your insights. - Hal cygcheck.out Description: cygcheck.out -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: [RSYNC] bad modify/change time set by rsync ?
Hi, --modify-window When comparing two timestamps, rsync treats the timestamps as being equal if they differ by no more than the modify-window value. This is normally 0 (for an exact match), but you may find it useful to set this to a larger value in some situations. In particular, when transferring to or from an MS Windows FAT filesystem (which represents times with a 2-second resolution), --modify-window=1 is useful (allowing times to differ by up to 1 second). Yes, (ex)FAT just doesn't have the resolution to represent some timestamps, specifically it can only record even seconds. Yes, that were also a guess at one moment in my testing time, but was discarded because, as I said, I had the very same problem on a ntfs partitions which have far greater precision. So is this normal ? fabrice -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Force "ls" to show .exe extension
Right now, if I have ~/bin/pdfcrop.exe, the command "ls ~/bin/pdfcrop" shows pdfcrop rather than pdfcrop.exe. Is there any way to force ls to show the full filename (including extension) if it matched the ls argument, even if the ls argument doesn't specify the extension? I read http://cygwin.com/cygwin-ug-net/using-specialnames.html, which helps explain the situation, but not a solution. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: [RSYNC] bad modify/change time set by rsync ?
LEGOND Fabrice writes: > You can see here that: > * the Modify/Change time differ from one second. Why ? Reading the man page would tell you why and what to do: --modify-window When comparing two timestamps, rsync treats the timestamps as being equal if they differ by no more than the modify-window value. This is normally 0 (for an exact match), but you may find it useful to set this to a larger value in some situations. In particular, when transferring to or from an MS Windows FAT filesystem (which represents times with a 2-second resolution), --modify-window=1 is useful (allowing times to differ by up to 1 second). > Or perhaps I miss something; If someone can help me. Yes, (ex)FAT just doesn't have the resolution to represent some timestamps, specifically it can only record even seconds. http://ntfs.com/exfat-time-stamp.htm Regards, Achim. -- +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+ SD adaptations for KORG EX-800 and Poly-800MkII V0.9: http://Synth.Stromeko.net/Downloads.html#KorgSDada -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
[RSYNC] bad modify/change time set by rsync ?
Hi, I use cygwin for a long time now. And I often use rsync to do incremental backup and mirroring of data. I noted several times in the past that rsync keeps copying some NOT MODIFIED files (always the same ones) again and again whereas it was working correctly for the vast majority of the remaining files. I see this problem for a long time ie on the cygwin 1.6/1.7 32 and 64 bit versions. It is difficult to reproduce the bug because I don't know what causing it. It can happen between remote (linux ssh - ext3/reiserfs/ext4 partitions) source and a target local NTFS, a remote source and a target local exFat, between 2 local ntfs partitions and between a local NTFS and a local target exFat (which has no right management). I have tried check permissions (because it was the most evident culprit to me) and tried to test many options combination of rsync (-X, -g, -no-g, -o, -no-o, -no-p, -p, -a, -t) and read numerous threads, but no luck here. Then, I checked the time and, rarely, on some files, the rsync does not seem to be able to correctly set the modify and change time of the target file. This problem is reproducible as soon as you have a bad file (but very few are): it will always be resynchronized whatever rsync options you set (even with/out -t/-a). Here an example of "malicious" file after a synchronization (with/out different combinations of -g -no-g -no-o -no-p -p -o, with/out some other obscure options, even after removing the target and its parents directory, but always with -t ...). stat of the source file (ntfs): File: '/cygdrive/e/Dev_Gen/android/apps/apk/Adobe_Flash_Player_10.1.120.1.apk' Size: 4544329 Blocks: 4480 IO Block: 65536 regular file Device: 54275946h/1411864902d Inode: 200264839444658 Links: 1 Access: (0600/-rw---) Uid: ( 1001/ xxx) Gid: ( 513/ Aucun) Access: 2014-11-05 22:18:30.0 +0100 Modify: 2014-05-19 14:52:43.0 +0200 Change: 2014-05-19 14:52:43.0 +0200 Birth: 2014-06-14 17:28:05.87000 +0200 stat of the destination file (exfat): File: '/cygdrive/f/backup_data/Dev_Gen/android/apps/apk/Adobe_Flash_Player_10.1.120.1.apk' Size: 4544329 Blocks: 4608 IO Block: 65536 regular file Device: e65d19c0h/3864861120d Inode: 2344130775972094377 Links: 1 Access: (0600/-rw---) Uid: ( 1001/ xxx) Gid: ( 513/ Aucun) Access: 2015-01-06 16:59:00.0 +0100 Modify: 2014-05-19 14:52:44.0 +0200 Change: 2014-05-19 14:52:44.0 +0200 Birth: 2014-10-26 14:37:20.57000 +0100 You can see here that: * the Modify/Change time differ from one second. Why ? * The Birth time value of the target file is impossible for me to understand but should not have any side-effect. However, is it a problem with cygwin ? * blocks should be irrelevant here because it can differ from FS to FS. Even after deleting the target, you get the same Modify/Change/Birth but the access time differ as the target file contains the (effective) date of the copy. On another "not malicious" file, you can see what seem the correct time in the two files. stat of the source file (ntfs): File: '/cygdrive/e/Dev_Gen/android/apps/apk/Adobe_Flash_Player_11.1.115.81.apk' Size: 4708736 Blocks: 4608 IO Block: 65536 regular file Device: 54275946h/1411864902d Inode: 14702358681444365420 Links: 1 Access: (0600/-rw---) Uid: ( 1001/ xxx) Gid: ( 513/ Aucun) Access: 2014-11-05 22:18:30.0 +0100 Modify: 2014-05-19 14:53:46.0 +0200 Change: 2014-05-19 14:53:46.0 +0200 Birth: 2014-06-14 17:28:06.96000 +0200 stat of the destination file (exfat): File: '/cygdrive/f/backup_data/Dev_Gen/android/apps/apk/Adobe_Flash_Player_11.1.115.81.apk' Size: 4708736 Blocks: 4608 IO Block: 65536 regular file Device: e65d19c0h/3864861120d Inode: 2811054816192366358 Links: 1 Access: (0600/-rw---) Uid: ( 1001/ xxx) Gid: ( 513/ Aucun) Access: 2015-01-06 16:59:02.0 +0100 Modify: 2014-05-19 14:53:46.0 +0200 Change: 2014-05-19 14:53:46.0 +0200 Birth: 2014-10-26 14:37:22.64000 +0100 Just to be clear, I also have numerous "not malicious" files with a different blocks size value. To conclude, AFTER 'touch'ing the source file (ie to the current time), the always copy behavior of rsync stop after one rsync as it should be. I really don't know if this is a cygwin problem (which can not correctly convert time from FS to FS, or a buggy setting function) or an internal rsync problem or a bad interaction between the two (most likely probable to me, rounding time ?). Or perhaps I miss something; If someone can help me. regards, fabrice -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: SFTP Cipher Mismatch
On 1/6/2015 2:47 PM, Chris Johnston wrote: Hello, I am attempting to set up Cygwin for Cisco’s Disaster Recovery System (9. 1.2.1-28) to run SFTP backups through. I have followed this guide that I found through online searches (https://supportforums.cisco.com/sites/default/files/legacy/0/3/2/41230-Cygwin%20Setup.pdf) and think I have everything set up correctly. I can use Filezilla through port 22 and the user account I set up through the Cygwin terminal to access the server I’d like to back up to. However when I attempt to set up the DRS on Cisco’s site I get an error message saying that “Update Failed: Unable to access SFTP server. Please ensure the Username and Password are correct.” I know what you’re thinking, “This is a Cisco Question, this guy sent to the wrong list”. Well I talked to Cisco TAC and after the run around they said to reach out to you folks because our CUCM is sending in aes-128-cbc, whereas Cygwin is replying back with aes-128-ctr, which they say is a mismatch and causing my problem. I checked my ssh_config and aes128-cbc is listed under Ciphers. Is there a way to change how Cygwin is replying? When searching I haven’t found a clear command that lets me change the default. Cisco’s suggestion was... To check if aes-128-cbc is enable , go to C:\cygwin\etc\sshd_config and check if the following line is there: Ciphers aes256-ctr,aes192-ctr,aes128-ctr,arcfour256,aes128-cbc aes-128-ctr is the first reported on my config eventually it is used as default; have you tried to put only a single line Ciphers aes128-cbc and check if that works ? If the line is there make sure you have aes128-cbc in there ...There was not a line specifying aes128-cbc in SSHD_Config, but as I understand it that means it would go to the default which was in the SSH_Config, which did include aes128-cbc. Cisco’s suggestion of copy and pasting that line in the sshd_config didn’t work. Any help would be greatly appreciated. All the best, cojohnst -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
SFTP Cipher Mismatch
Hello, I am attempting to set up Cygwin for Cisco’s Disaster Recovery System (9. 1.2.1-28) to run SFTP backups through. I have followed this guide that I found through online searches (https://supportforums.cisco.com/sites/default/files/legacy/0/3/2/41230-Cygwin%20Setup.pdf) and think I have everything set up correctly. I can use Filezilla through port 22 and the user account I set up through the Cygwin terminal to access the server I’d like to back up to. However when I attempt to set up the DRS on Cisco’s site I get an error message saying that “Update Failed: Unable to access SFTP server. Please ensure the Username and Password are correct.” I know what you’re thinking, “This is a Cisco Question, this guy sent to the wrong list”. Well I talked to Cisco TAC and after the run around they said to reach out to you folks because our CUCM is sending in aes-128-cbc, whereas Cygwin is replying back with aes-128-ctr, which they say is a mismatch and causing my problem. I checked my ssh_config and aes128-cbc is listed under Ciphers. Is there a way to change how Cygwin is replying? When searching I haven’t found a clear command that lets me change the default. Cisco’s suggestion was... To check if aes-128-cbc is enable , go to C:\cygwin\etc\sshd_config and check if the following line is there: Ciphers aes256-ctr,aes192-ctr,aes128-ctr,arcfour256,aes128-cbc If the line is there make sure you have aes128-cbc in there ...There was not a line specifying aes128-cbc in SSHD_Config, but as I understand it that means it would go to the default which was in the SSH_Config, which did include aes128-cbc. Cisco’s suggestion of copy and pasting that line in the sshd_config didn’t work. Any help would be greatly appreciated. All the best, cojohnst -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
