Re: [TESTERS needed] New POSIX permission handling
Hi Ismail, On Apr 16 09:48, Ismail Donmez wrote: Corinna Vinschen-2 wrote On Apr 16 09:09, Ismail Donmez wrote: [~/src/mutt] chmod +x config.status chmod: changing permissions of ‘config.status’: Permission denied This is really surprising. [~/src/mutt] icacls config.status config.status NULL SID:(DENY)(Rc,S,REA,X,DC) UX31A\ismail:(F) UX31A\ismail:(RX) NT AUTHORITY\SYSTEM:(RX,W,DC) BUILTIN\Administrators:(RX,W,DC) Everyone:(RX) Successfully processed 1 files; Failed processing 0 files [~/src/mutt] icacls . . NULL SID:(DENY)(Rc,S,REA,X,DC) UX31A\ismail:(F) UX31A\ismail:(RX) NT AUTHORITY\SYSTEM:(RX,W,DC) BUILTIN\Administrators:(RX,W,DC) Everyone:(RX) NULL SID:(OI)(CI)(IO)(DENY)(Rc,S,REA,X,DC) CREATOR OWNER:(OI)(CI)(IO)(F) CREATOR GROUP:(OI)(CI)(IO)(RX) NT AUTHORITY\SYSTEM:(OI)(CI)(IO)(RX,W,DC) BUILTIN\Administrators:(OI)(CI)(IO)(RX,W,DC) Everyone:(OI)(CI)(IO)(RX) I created the exact same scenario with the exact same ACLs, just with my own account as user and group, and I can't reproduce a Permission denied. chmod simply works for me. Weird. While looking into this I found some other problems in this owner==group scenario, so I applied a few patches. I'm just creating new developer snapshots and a 2.0.0-0.6 test release. It would be nice if you could update, test again, and if the problem still occurs, I think I'll need an strace of the above chmod call as in: $ strace -o chmod.trace chmod +x config.status This combined with the icacls of the ACLs of config.status and the parent dir, just to be sure the new Cygwin created the same ACLs. Thanks in advance, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpYraRDSUoPl.pgp Description: PGP signature
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On Apr 16 14:14, Ian Lambert wrote: It might be interesting to check the files. Don't delete them, they may come in handy for forensics. If you don't mind and they don;t contain anything too private, you might even just copy them here. I can post the good and bad ones tomorrow. Similar problem today. Previous 2.0.0-0.x worked OK, and going back to 1.7.35-1 works OK. No admin rights. No passwd or group files. nsswitch.conf all commented out. Some excerpts follow. Try the -0.6 I'm just uploading. This should work again since it checks the PSIDs for NULL before calling this crashing OS function. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpvBE1ZhGYCg.pgp Description: PGP signature
[newlib-cygwin] Fix broken PSID problem on cygheap in account handling
https://sourceware.org/git/gitweb.cgi?p=newlib-cygwin.git;h=ecc6cfeb7e6d6bb1be8cca563d22c6f8c2c6add1
commit ecc6cfeb7e6d6bb1be8cca563d22c6f8c2c6add1
Author: Corinna Vinschen cori...@vinschen.de
Date: Fri Apr 17 11:56:15 2015 +0200
Fix broken PSID problem on cygheap in account handling
* pwdgrp.h: Add comment to explain below change.
(struct pg_pwd): Convert sid member to BYTE array.
(struct pg_grp): Ditto.
* grp.cc (pwdgrp::parse_group): Accommodate above change.
* passwd.cc (pwdgrp::parse_passwd): Ditto.
Signed-off-by: Corinna Vinschen cori...@vinschen.de
Diff:
---
winsup/cygwin/ChangeLog | 8
winsup/cygwin/grp.cc| 4 +++-
winsup/cygwin/passwd.cc | 4 +++-
winsup/cygwin/pwdgrp.h | 8 ++--
4 files changed, 20 insertions(+), 4 deletions(-)
diff --git a/winsup/cygwin/ChangeLog b/winsup/cygwin/ChangeLog
index 2ecafa0..a340261 100644
--- a/winsup/cygwin/ChangeLog
+++ b/winsup/cygwin/ChangeLog
@@ -1,3 +1,11 @@
+2015-04-17 Corinna Vinschen cori...@vinschen.de
+
+ * pwdgrp.h: Add comment to explain below change.
+ (struct pg_pwd): Convert sid member to BYTE array.
+ (struct pg_grp): Ditto.
+ * grp.cc (pwdgrp::parse_group): Accommodate above change.
+ * passwd.cc (pwdgrp::parse_passwd): Ditto.
+
2015-04-16 Corinna Vinschen cori...@vinschen.de
* sec_acl.cc (set_posix_access): Move merging group perms into owner
diff --git a/winsup/cygwin/grp.cc b/winsup/cygwin/grp.cc
index 40e1ca7..f850210 100644
--- a/winsup/cygwin/grp.cc
+++ b/winsup/cygwin/grp.cc
@@ -47,7 +47,9 @@ pwdgrp::parse_group ()
return false;
/* Don't generate gr_mem entries. */
grp.g.gr_mem = null_ptr;
- grp.sid.getfromgr_passwd (grp.g);
+ cygsid csid;
+ csid.getfromgr_passwd (grp.g);
+ RtlCopySid (SECURITY_MAX_SID_SIZE, grp.sid, csid);
return true;
}
diff --git a/winsup/cygwin/passwd.cc b/winsup/cygwin/passwd.cc
index 7493aa4..54e429c 100644
--- a/winsup/cygwin/passwd.cc
+++ b/winsup/cygwin/passwd.cc
@@ -40,7 +40,9 @@ pwdgrp::parse_passwd ()
res.p.pw_gecos = next_str (':');
res.p.pw_dir = next_str (':');
res.p.pw_shell = next_str (':');
- res.sid.getfrompw_gecos (res.p);
+ cygsid csid;
+ csid.getfrompw_gecos (res.p);
+ RtlCopySid (SECURITY_MAX_SID_SIZE, res.sid, csid);
/* lptr points to the \0 after pw_shell. Increment by one to get the correct
required buffer len in getpw_cp. */
res.len = lptr - res.p.pw_name + 1;
diff --git a/winsup/cygwin/pwdgrp.h b/winsup/cygwin/pwdgrp.h
index a64cb0f..80c54c7 100644
--- a/winsup/cygwin/pwdgrp.h
+++ b/winsup/cygwin/pwdgrp.h
@@ -38,17 +38,21 @@ void *setgrent_filtered (int enums, PCWSTR enum_tdoms);
void *getgrent_filtered (void *gr);
void endgrent_filtered (void *gr);
+/* NOTE: The below sid members were cygsid's originally. Don't do that.
+ cygsid's are pointer based. When adding new entries to the passwd or
+ group caches, a crealloc call potenitally moves the entries and then
+ the cygsid pointers point into neverneverland. */
struct pg_pwd
{
struct passwd p;
- cygsid sid;
+ BYTE sid[SECURITY_MAX_SID_SIZE];
size_t len;
};
struct pg_grp
{
struct group g;
- cygsid sid;
+ BYTE sid[SECURITY_MAX_SID_SIZE];
size_t len;
};
[newlib-cygwin] Fix broken SID in passwd/group entry for unkown account
https://sourceware.org/git/gitweb.cgi?p=newlib-cygwin.git;h=027506709a0a821cdaa9a6ff0f07fb300dad
commit 027506709a0a821cdaa9a6ff0f07fb300dad
Author: Corinna Vinschen cori...@vinschen.de
Date: Fri Apr 17 11:57:57 2015 +0200
Fix broken SID in passwd/group entry for unkown account
* uinfo.cc (pwdgrp::fetch_account_from_windows): Always revert SID
subauth count after checking for known domain.
Signed-off-by: Corinna Vinschen cori...@vinschen.de
Diff:
---
winsup/cygwin/ChangeLog | 5 +
winsup/cygwin/uinfo.cc | 2 +-
2 files changed, 6 insertions(+), 1 deletion(-)
diff --git a/winsup/cygwin/ChangeLog b/winsup/cygwin/ChangeLog
index a340261..fb66f39 100644
--- a/winsup/cygwin/ChangeLog
+++ b/winsup/cygwin/ChangeLog
@@ -1,5 +1,10 @@
2015-04-17 Corinna Vinschen cori...@vinschen.de
+ * uinfo.cc (pwdgrp::fetch_account_from_windows): Always revert SID
+ subauth count after checking for known domain.
+
+2015-04-17 Corinna Vinschen cori...@vinschen.de
+
* pwdgrp.h: Add comment to explain below change.
(struct pg_pwd): Convert sid member to BYTE array.
(struct pg_grp): Ditto.
diff --git a/winsup/cygwin/uinfo.cc b/winsup/cygwin/uinfo.cc
index 6186327..bc5f814 100644
--- a/winsup/cygwin/uinfo.cc
+++ b/winsup/cygwin/uinfo.cc
@@ -2475,10 +2475,10 @@ pwdgrp::fetch_account_from_windows (fetch_user_arg_t
arg, cyg_ldap *pldap)
posix_offset = fetch_posix_offset (td, loc_ldap);
break;
}
+ sid_sub_auth_count (sid) = sid_sub_auth_count (sid) + 1;
}
if (domain)
{
- sid_sub_auth_count (sid) = sid_sub_auth_count (sid) + 1;
wcscpy (dom, domain);
__small_swprintf (name = namebuf, L%W(%u),
is_group () ? LGroup : LUser,
Re: [TESTERS needed] New POSIX permission handling
On Apr 17 09:30, Corinna Vinschen wrote: Hi Ismail, On Apr 16 09:48, Ismail Donmez wrote: Corinna Vinschen-2 wrote On Apr 16 09:09, Ismail Donmez wrote: [~/src/mutt] chmod +x config.status chmod: changing permissions of ‘config.status’: Permission denied This is really surprising. [~/src/mutt] icacls config.status config.status NULL SID:(DENY)(Rc,S,REA,X,DC) UX31A\ismail:(F) UX31A\ismail:(RX) NT AUTHORITY\SYSTEM:(RX,W,DC) BUILTIN\Administrators:(RX,W,DC) Everyone:(RX) Successfully processed 1 files; Failed processing 0 files [~/src/mutt] icacls . . NULL SID:(DENY)(Rc,S,REA,X,DC) UX31A\ismail:(F) UX31A\ismail:(RX) NT AUTHORITY\SYSTEM:(RX,W,DC) BUILTIN\Administrators:(RX,W,DC) Everyone:(RX) NULL SID:(OI)(CI)(IO)(DENY)(Rc,S,REA,X,DC) CREATOR OWNER:(OI)(CI)(IO)(F) CREATOR GROUP:(OI)(CI)(IO)(RX) NT AUTHORITY\SYSTEM:(OI)(CI)(IO)(RX,W,DC) BUILTIN\Administrators:(OI)(CI)(IO)(RX,W,DC) Everyone:(OI)(CI)(IO)(RX) I created the exact same scenario with the exact same ACLs, just with my own account as user and group, and I can't reproduce a Permission denied. chmod simply works for me. Weird. While looking into this I found some other problems in this owner==group scenario, so I applied a few patches. I'm just creating new developer snapshots and a 2.0.0-0.6 test release. It would be nice if you could update, test again, and if the problem still occurs, I think I'll need an strace of the above chmod call as in: $ strace -o chmod.trace chmod +x config.status This combined with the icacls of the ACLs of config.status and the parent dir, just to be sure the new Cygwin created the same ACLs. I think I found the culprit. I'll uploade a -0.7 test release in the next hour or so. Thanks for testing, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpgSkwvT2e5G.pgp Description: PGP signature
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On Apr 16 21:49, Denis Excoffier wrote: On 2015-04-16 à 18:43, Corinna Vinschen wrote: On Apr 16 18:21, Corinna Vinschen wrote: On Apr 16 08:17, Jim Reisert AD1C wrote: I am unable to start Cywin/X X-server 1.17.1 with this version. Previous releases of 2.0.0.x were OK. I had to revert to 1.7.35-1 for the time being. Other than updating to 2.0.0.5, I also installed the April 2015 Patch Tuesday updates from Microsoft. I don't know if the two are related. Windows 7 Home Premium, 64-bit Exception: STATUS_ACCESS_VIOLATION at eip=77C50F8A eax= ebx=612D67B0 ecx=1000 edx=612D2648 esi= edi=0028C790 ebp=0028C608 esp=0028C604 program=C:\Cygwin\bin\XWin.exe, pid 1660, thread main cs=0023 ds=002B es=002B fs=0053 gs=002B ss=002B Stack trace: Frame Function Args 0028C608 77C50F8A (, 612D2648, , 612D67B0) 0028C738 610CDA1F (43FF, , , 80012428) On second thought, if I can trust the args output, that would be an fchmod(0,0). If there's no uid or gid 0, which there isn't unless you explicitely created them in the passwd/group files, the uid and gid have no SID connected to. This may be the culprit here. I could add an extra check which refuses to change permissions if the account's SID can't be found, but since this occurs very deep in the call stack, the error message might be pretty vapid. Alternatively I just let this slip through and you might wonder why the group hasn't changed in this case. I added a change to this effect, but it occuurs to me that this may be really just a missing test if the uid and gid values are backed by a real Windows account. It seems better to return EPERM here. I applied the patch indicated (see in https://cygwin.com/ml/cygwin-cvs/2015-q2/msg00033.html) and X11 now works back again. Thank you 1000 times. Thanks for the early feedback. Still, I changed it for -0.6 again, because that's what had happened in the old code, too: If the uid or gid can't be resolved to a Windows SID, the function fails now with ENOENT. That's a non-POSIX extension because the problem just doesn't exist on real POSIX systems. Please give it a try. Thanks, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgppWLEz7Jr_l.pgp Description: PGP signature
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On Apr 16 12:53, Bryan Berns wrote: On Thu, Apr 16, 2015 at 10:17 AM, Jim Reisert AD1C jjreis...@alum.mit.edu wrote: I am unable to start Cywin/X X-server 1.17.1 with this version. Previous releases of 2.0.0.x were OK. I had to revert to 1.7.35-1 for the time being. Other than updating to 2.0.0.5, I also installed the April 2015 Patch Tuesday updates from Microsoft. I don't know if the two are related. Windows 7 Home Premium, 64-bit Exception: STATUS_ACCESS_VIOLATION at eip=77C50F8A eax= ebx=612D67B0 ecx=1000 edx=612D2648 esi= edi=0028C790 ebp=0028C608 esp=0028C604 program=C:\Cygwin\bin\XWin.exe, pid 1660, thread main cs=0023 ds=002B es=002B fs=0053 gs=002B ss=002B Stack trace: Frame Function Args 0028C608 77C50F8A (, 612D2648, , 612D67B0) 0028C738 610CDA1F (43FF, , , 80012428) 0028C7B8 61047198 (, 72483F24, 75604227, 0254) 0028C7F8 610F629D (0001, , , 75623912) -- Jim Reisert AD1C, jjreis...@alum.mit.edu, http://www.ad1c.us I've actually had problem building Cygwin (1.7.35 or 2.x source) on Cygwin 2.x using Windows. The make errors out with a permissions denied when setting permissions (chmod +x) on config.status in builddir/etc. I was able to produce it on three different, freshly built copies of Windows (no BLODA at all). Only had Cygwin plus the build essentials (gcc, ming, xmlto, diffutils, textinfo, cocom) installed. Only tried with 32-bit. The super wacky thing is that if I rearrange the unrelated contents of the make file that fuels the whole process, the problem goes away. If I downgrade the core back to 1.7.35, the problem goes away. It almost seems like there might an uninitialized memory issue in the code path executed in chmod(). I apologize for the lack of debugging on my part, but I just wanted to throw this out there in case a) it could be related to this issue or b) anyone else has seen this. This is more in line with what Ismail reported in https://cygwin.com/ml/cygwin/2015-04/msg00365.html And I finally managed to reproduce it. It works on the command line and only fails during `make' for me. Investigating now... That said, I'm making progress in debugging my Unknown Group issue described in another chain; hope to report out on that this weekend when I have a little more time to play. Cool, thanks. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpbKE9knd1nc.pgp Description: PGP signature
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On Apr 17 10:16, Corinna Vinschen wrote: On Apr 16 12:53, Bryan Berns wrote: On Thu, Apr 16, 2015 at 10:17 AM, Jim Reisert AD1C jjreis...@alum.mit.edu wrote: I am unable to start Cywin/X X-server 1.17.1 with this version. Previous releases of 2.0.0.x were OK. I had to revert to 1.7.35-1 for the time being. Other than updating to 2.0.0.5, I also installed the April 2015 Patch Tuesday updates from Microsoft. I don't know if the two are related. Windows 7 Home Premium, 64-bit Exception: STATUS_ACCESS_VIOLATION at eip=77C50F8A eax= ebx=612D67B0 ecx=1000 edx=612D2648 esi= edi=0028C790 ebp=0028C608 esp=0028C604 program=C:\Cygwin\bin\XWin.exe, pid 1660, thread main cs=0023 ds=002B es=002B fs=0053 gs=002B ss=002B Stack trace: Frame Function Args 0028C608 77C50F8A (, 612D2648, , 612D67B0) 0028C738 610CDA1F (43FF, , , 80012428) 0028C7B8 61047198 (, 72483F24, 75604227, 0254) 0028C7F8 610F629D (0001, , , 75623912) -- Jim Reisert AD1C, jjreis...@alum.mit.edu, http://www.ad1c.us I've actually had problem building Cygwin (1.7.35 or 2.x source) on Cygwin 2.x using Windows. The make errors out with a permissions denied when setting permissions (chmod +x) on config.status in builddir/etc. I was able to produce it on three different, freshly built copies of Windows (no BLODA at all). Only had Cygwin plus the build essentials (gcc, ming, xmlto, diffutils, textinfo, cocom) installed. Only tried with 32-bit. The super wacky thing is that if I rearrange the unrelated contents of the make file that fuels the whole process, the problem goes away. If I downgrade the core back to 1.7.35, the problem goes away. It almost seems like there might an uninitialized memory issue in the code path executed in chmod(). I apologize for the lack of debugging on my part, but I just wanted to throw this out there in case a) it could be related to this issue or b) anyone else has seen this. This is more in line with what Ismail reported in https://cygwin.com/ml/cygwin/2015-04/msg00365.html And I finally managed to reproduce it. It works on the command line and only fails during `make' for me. Investigating now... I think I found the culprit. I'll uploade a -0.7 test release in the next hour or so. Thanks for testing, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpe6mpRBXAIT.pgp Description: PGP signature
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On Apr 16 15:20, Jim Reisert AD1C wrote: I would suggest not to use passwd and group files anymore, unless you're in a situation with no connection to the account DB (e.g. AD member machine without connection to the DC). If you mean remove the /etc/passwd and /etc/group files, that didn't work for me. Xwin crashed the same way. That should be fixed in 2.0.0-0.6 which I'm just copying to the release area. mkpasswd isn't an option for me, everyone in the company will be added to the password file (I tried it). Non-starter! As I said, just don't use them. But if you need it for some reason, try $ mkpasswd -l -c $ mkgroup -l -c That restricts the outout to the local accoutns and your own user and primary group. That's what setup calls. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpgpuz_qSFEH.pgp Description: PGP signature
Re: Cygwin hangs up if several keys are typed during outputting a lot of texts.
Hi Corinna, On Thu, 16 Apr 2015 11:05:33 +0200 Corinna Vinschen corinna-cyg...@cygwin.com wrote: Ok, but... this is a really big patch and it complicates the pty code even more. Is there really no other option as far as the TCSADRAIN problem is concerned? What strikes me as weird is that neither fhandler_pty_slave::tcsetattr nor fhandler_pty_master::tcsetattr give a damn for the optional_actions parameter. They simply overwrite the tc settings. So I'm wondering, wouldn't it be possible to add code to the tcsetattr implementation instead, so that TCSADRAIN/TCSAFLUSH are honored and than only have one place for OPOST handling? I also think the patch was a big deal. However, I did not have any other good idea. Anyway, I have worked out another solution. Please find a patch attached. What do you think of this one? OPOST code has been now completely moved back to master side as with original implementation. With this patch, tcsetattr() waits until master reads all data in pipe before new attributes are applied to preserve the order between write() and tcsetattr(). However, there is a potential risk in which tcsetattr() can be blocked if master stops reading pipe, even though I can not imagine such a likely situation. Moreover, it is necessary to duplicate io_handle from master-side to slave-side to allow slave to peek master pipe. It is not smart enough, I suppose. ChageLog is as follows. 2015-04-17 Takashi Yano takashi.y...@nifty.ne.jp * fhandler.h (fhandler_pty_common): Remove function process_opost_output (). Since processing regarding OPOST is moved back to master side, this function is needed no more. (class fhandler_pty_slave): Add variable 'from_slave' for checking whether master input pipe is empty. This used by flandler_pty_slave ::tcsetattr(). (class fhandler_pty_master): Move variable 'column' for handling ONOCR from class tty to class fhandler_pty_master. Revive variable 'need_nl' which is used by OPOST-processing in master side. * fhandler_tty.cc (struct pipe_reply): Add member 'from_slave'. (fhandler_pty_master::doecho): Call WriteFile() instead of fhandler_pty_common::process_opost_output(). (fhandler_pty_master::process_slave_output): Move OPOST-processing back into this function. (fhandler_pty_slave::fhandler_pty_slave): Initialize 'from_slave'. (fhandler_pty_slave::open): Duplicate handle regarding 'from_slave'. (fhandler_pty_slave::close): Close handle 'from_slave'. (fhandler_pty_slave::write): Call WriteFile() instead of fhandler_pty_common::process_opost_output(). (fhandler_pty_slave::tcsetattr): Wait until master reads all data from slave. (fhandler_pty_slave::fhandler_pty_slave): Initialize 'column' and 'need_nl'. (fhandler_pty_master::pty_master_thread): Add code to duplicate handle regarding 'from_slave'. (fhandler_pty_master::setup): Set io_handle to tty::_from_slave. (fhandler_pty_common::process_opost_output): Remove this function. * select.cc (peek_pipe): Revive the code checking 'need_nl'. * tty.cc (tty::init): Remove initialization for 'column'. * tty.h (class tty): Add variable '_from_slave'. Add two functions from_slave() and set_from_slave() regarding '_from_master'. Remove variable 'column'. It is moved to class fhandler_pty_master. -- Takashi Yano takashi.y...@nifty.ne.jp cygwin.patch.20150417-2 Description: Binary data -- 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: TEST RELEASE: Cygwin 2.0.0-0.7
New 2.0.0-0.7 test release: I'm happy to report that Xwin now starts up straight away, even with missing /etc/passwd and /etc/group files. Thanks for fixing this! -- Jim Reisert AD1C, jjreis...@alum.mit.edu, http://www.ad1c.us -- 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 hangs up if several keys are typed during outputting a lot of texts.
On Apr 17 14:10, Corinna Vinschen wrote:
On Apr 17 20:27, Takashi Yano wrote:
@@ -868,6 +980,9 @@ fhandler_pty_slave::tcgetattr (struct termios *t)
int
fhandler_pty_slave::tcsetattr (int, const struct termios *t)
{
+ DWORD n;
+ while (::bytes_available (n, from_slave) n)
+cygwait (10);
acquire_output_mutex (INFINITE);
get_ttyp ()-ti = *t;
release_output_mutex ();
Shouldn't this loop be skipped in TCSANOW mode?
OPOST code has been now completely moved back to master side as
with original implementation.
With this patch, tcsetattr() waits until master reads all data
in pipe before new attributes are applied to preserve the order
between write() and tcsetattr().
However, there is a potential risk in which tcsetattr() can be
blocked if master stops reading pipe, even though I can not imagine
such a likely situation.
Yeah, but it is a busy wait. Hmm. Also, on second thought, the above
loop checks for bytes_available every time and changes n. So, if
another slave writes lots of data, wouldn't the slave calling tcsetattr
starve?
IIUC, what you'd really like to know is something else. It's not about
having n 0 bytes in the pipe, but on calling tcsetattr, you'd like to
know how much bytes are in the pipe at this very moment, and then you'd
overwrite the termios info as soon as these n bytes are written. That
sounds pretty different to me.
It would be cool if the slave-side tcsetattr only transmits the
optional_actions and the new termios content to the master, and the
master keeps it stashed away together with the number of bytes in the
pipe right now. Then in, say, process_slave_output, it checks if the
precondition is fulfilled and only then overwrites its termios
structure.
Off the top of my head I'm not sure how feasible this is. One way to do
that *may* be to send the info in the normal write pipe to the master.
What we need then would be a method to identify such a tcsetattr packet
in the input stream.
Other ideas?
Maybe your idea to introduce a second pipe wasn't that bad after all...
So, instead of using it for Cygwin slave I/O (which makes me cringe a
bit), it could be used as a command channel to the master. Since only
Cygwin executables would understand this concept anyway, it's ok that
native applications don't know about it. This pipe could then be used
to transmit tcsetattr info to the master, and the master could apply the
change when it sees fit. Maybe we even realize it could be used for
something else in future. Ctrl-S/Ctrl-Q processing might be nice, say.
What do you think?
I'm sorry if I let you jump through hoops here, but I'm not really
sure about the best solution either :}
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Maintainer cygwin AT cygwin DOT com
Red Hat
pgpb_IFuYzXpC.pgp
Description: PGP signature
Re: TEST RELEASE: Cygwin 2.0.0-0.7
On Apr 17 06:15, Jim Reisert AD1C wrote: New 2.0.0-0.7 test release: I'm happy to report that Xwin now starts up straight away, even with missing /etc/passwd and /etc/group files. Thanks for fixing this! Sure, thanks for your feedback! Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpG82d_JBGSF.pgp Description: PGP signature
Re: TEST RELEASE: Cygwin 2.0.0-0.7
On Apr 17 06:27, Jim Reisert AD1C wrote: There's still something funky going on. ssh won't recognize my ~/.ssh/config file. With my existing passwd and group files, I get this error (notice that the warned permissions are different that what ls reports): -rw---+ 1 Jim Reisert None 205 Apr 17 06:19 .ssh/config [JJR:~] $ ssh ad1c @@@ @ WARNING: UNPROTECTED PRIVATE KEY FILE! @ @@@ Permissions 0644 for '/cygdrive/D/home/.ssh/id_rsa' are too open. ^^ Another file? It is recommended that your private key files are NOT accessible by others. This private key will be ignored. Load key /cygdrive/D/home/.ssh/id_rsa: bad permissions If I move both passwd and group out-of-the-way, the config file is not recognized at all! -rw---+ 1 Jim Reisert None 205 Apr 17 06:19 .ssh/config [JJR:~] $ ssh ad1c ssh: Could not resolve hostname ad1c: Name or service not known I don't understand the reason for this message. It does not exactly point to a problem with the config file. I think I need more explanation here and a bit of detective work from your side... Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpxKHZNfvIRo.pgp Description: PGP signature
Re: TEST RELEASE: Cygwin 2.0.0-0.7
There's still something funky going on. ssh won't recognize my ~/.ssh/config file. With my existing passwd and group files, I get this error (notice that the warned permissions are different that what ls reports): -rw---+ 1 Jim Reisert None 205 Apr 17 06:19 .ssh/config [JJR:~] $ ssh ad1c @@@ @ WARNING: UNPROTECTED PRIVATE KEY FILE! @ @@@ Permissions 0644 for '/cygdrive/D/home/.ssh/id_rsa' are too open. It is recommended that your private key files are NOT accessible by others. This private key will be ignored. Load key /cygdrive/D/home/.ssh/id_rsa: bad permissions If I move both passwd and group out-of-the-way, the config file is not recognized at all! -rw---+ 1 Jim Reisert None 205 Apr 17 06:19 .ssh/config [JJR:~] $ ssh ad1c ssh: Could not resolve hostname ad1c: Name or service not known -- Jim Reisert AD1C, jjreis...@alum.mit.edu, http://www.ad1c.us -- 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: TEST RELEASE: Cygwin 2.0.0-0.7
I finally understand what happened. In my /etc/passwd, I have configured a different home directory for myself. Without that file, it goes to /home/Jim Reisert which does not exist! Is there a way to change my default home directory without using /etc/passwd, other than creating a symbolic link inside of /home? - Jim -- Jim Reisert AD1C, jjreis...@alum.mit.edu, http://www.ad1c.us -- 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: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On 16/04/2015 17:21, Corinna Vinschen wrote: Btw., it would be nice to know why XWin tries to chown a file. Jon? This seems to be in trans_mkdir() in libxtrans [1], which tries to make sure that /tmp/.X11-unix (the directory in which the server's unix-domain sockets are created) is owned by uid 0, gid 0. [1] http://cgit.freedesktop.org/xorg/lib/libxtrans/tree/Xtransutil.c#n591 $ gdb XWin [...] Reading symbols from XWin...Reading symbols from /usr/lib/debug//usr/bin/XWin.exe.dbg...done. done. (gdb) b fchown Breakpoint 1 at 0x10057aa80 (gdb) r Starting program: /usr/bin/XWin Breakpoint 1, 0x00010057aa80 in fchown () (gdb) bt #0 0x00010057aa80 in fchown () #1 0x000100573bac in trans_mkdir (mode=1023, path=0x1005b47c8 __func__.16196+2499 /tmp/.X11-unix) at /usr/include/X11/Xtrans/Xtransutil.c:591 [...] -- 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: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
Try the -0.6 I'm just uploading. This should work again since it checks the PSIDs for NULL before calling this crashing OS function. Indeed, it works again. Thanks! And ssh works too. :) It still gives this error message, but it was only noticed because it was the last line of output: winMultiWindowXMsgProcErrorHandler - ERROR: BadMatch (invalid parameter attributes) -- 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: TEST RELEASE: Cygwin 2.0.0-0.7
On Apr 17 15:16, Corinna Vinschen wrote: On Apr 17 06:46, Jim Reisert AD1C wrote: I finally understand what happened. In my /etc/passwd, I have configured a different home directory for myself. Without that file, it goes to /home/Jim Reisert which does not exist! Is there a way to change my default home directory without using /etc/passwd, other than creating a symbolic link inside of /home? That's what is explained in https://cygwin.com/cygwin-ug-net/ntsec.html The chapters you should be looking for are https://cygwin.com/cygwin-ug-net/ntsec.html#ntsec-mapping-nsswitch-home E.g. this setting in /etc/nsswitch.conf would do the trick if you're the only user (but careful with sshd!): db_home: /cygdrive/d/home Or perhaps https://cygwin.com/cygwin-ug-net/ntsec.html#ntsec-mapping-nsswitch-desc Set db_home: desc and change the description field in your SAM according to the above document. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgptASafg3ZTk.pgp Description: PGP signature
Re: TEST RELEASE: Cygwin 2.0.0-0.7
On Apr 17 06:46, Jim Reisert AD1C wrote: I finally understand what happened. In my /etc/passwd, I have configured a different home directory for myself. Without that file, it goes to /home/Jim Reisert which does not exist! Is there a way to change my default home directory without using /etc/passwd, other than creating a symbolic link inside of /home? That's what is explained in https://cygwin.com/cygwin-ug-net/ntsec.html The chapters you should be looking for are https://cygwin.com/cygwin-ug-net/ntsec.html#ntsec-mapping-nsswitch-home E.g. this setting in /etc/nsswitch.conf would do the trick if you're the only user (but careful with sshd!): Or perhaps https://cygwin.com/cygwin-ug-net/ntsec.html#ntsec-mapping-nsswitch-desc Set db_home: desc and change the description field in your SAM according to the above document. HTH, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpKujJ8fLeGQ.pgp Description: PGP signature
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On Apr 17 06:57, Bryan Berns wrote: On Fri, Apr 17, 2015 at 6:07 AM, Corinna Vinschen corinna-cyg...@cygwin.com wrote: On Apr 17 10:16, Corinna Vinschen wrote: On Apr 16 12:53, Bryan Berns wrote: I've actually had problem building Cygwin (1.7.35 or 2.x source) on Cygwin 2.x using Windows. The make errors out with a permissions denied when setting permissions (chmod +x) on config.status in builddir/etc. I was able to produce it on three different, freshly built copies of Windows (no BLODA at all). Only had Cygwin plus the build essentials (gcc, ming, xmlto, diffutils, textinfo, cocom) installed. Only tried with 32-bit. The super wacky thing is that if I rearrange the unrelated contents of the make file that fuels the whole process, the problem goes away. If I downgrade the core back to 1.7.35, the problem goes away. It almost seems like there might an uninitialized memory issue in the code path executed in chmod(). I apologize for the lack of debugging on my part, but I just wanted to throw this out there in case a) it could be related to this issue or b) anyone else has seen this. This is more in line with what Ismail reported in https://cygwin.com/ml/cygwin/2015-04/msg00365.html And I finally managed to reproduce it. It works on the command line and only fails during `make' for me. Investigating now... I think I found the culprit. I'll uploade a -0.7 test release in the next hour or so. Thanks for testing, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat On my way to work now and probably will be offline most of the day, but just wanted to let you know that I did a quick rip of the DLL file from cygwin-2.0.0-0.7.tar.xz on the sourceware.org mirror and make / chmod seems happier now. Nice! :-) Cool, thanks for testing! Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpabryP7CKGM.pgp Description: PGP signature
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On Apr 17 12:29, Jon TURNEY wrote: On 16/04/2015 17:21, Corinna Vinschen wrote: Btw., it would be nice to know why XWin tries to chown a file. Jon? This seems to be in trans_mkdir() in libxtrans [1], which tries to make sure that /tmp/.X11-unix (the directory in which the server's unix-domain sockets are created) is owned by uid 0, gid 0. Yep, that is in line with the observation. Thanks for confirming. Such an fchown is bound to fail, of course, but it did so before and XWin didn't gripe, so I assume it doesn't check the return value. Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpjKluBXAyR0.pgp Description: PGP signature
TEST RELEASE: Cygwin 2.0.0-0.7
Hi Cygwin friends and users, New 2.0.0-0.7 test release: - Improved setfacl tool. It now handles mask recomputation just like the Linux tool. -d option renamed to -x (but -d is still accepted for backward compat). New -n,--no-mask and --mask options. - Improved getfacl tool. It now prints effective permissions just like the Linux tool. --noname optioned renamed to --numeric (but --noname is still accepted for backward compat). New -c,--omit-header, -e,--all-effective, and -E,--no-effective options. - Fix a bug in account handling. In certain scenarios a PSID pointer was pointing into a buffer space which got moved away due to a call to realloc. This bug seems to be the problem of the reported Permission denied problems when trying to chmod. - Add a missing check for uid/gid values which can't be resolved to Windows accounts. chmod now returns EINVAL now in this case, just as the pre 2.0 DLLs. - I introduced a change in chmod behaviour which is not exactly in line with POSIX 1003.1e draft 17: If you have a non-trivial ACL with secondary accounts and thus a mask value, chmod is supposed to change only the mask, not the permissions of the primary group. However, if the primary group has few permissions to begin with, the result is really surprising. ls -l would, e.g., show read/write perms for the group, but the group might still have only read perms. Personally I find this chmod behaviour really, really bad, so I took the liberty to change it in a way which gives a much less surprising result: If you call chmod on a non-trivial ACL, the group permissions will be used for the primary group and the mask. I hope that this is working out ok. - Fix a bug when reading DENAY ACEs of accounts with multiple entries (typically in the owner == group case). - Fix how group perms are merged into owner perms in owner == group case. - Fix a bug in SID handling which may result in broken SID info in passwd/group entries of unknown accounts. Other than that... The important change in this release is the POSIX permission handling change, a rewrite of the underlying routines reading and creating Windows ACLs following POSIX permission rules and POSIX ACL creating rules per POSIX 1003.1e draft 17, as on Linux. For a description of POSIX ACLs, see http://linux.die.net/man/5/acl All changes in this release so far: === - New, unified implementation of POSIX permission and ACL handling. The new ACLs now store the POSIX ACL MASK/CLASS_OBJ permission mask, and they allow to inherit the S_ISGID bit. ACL inheritance now really works as desired, in a limited, but theoretically equivalent fashion even for non-Cygwin processes. To accommodate Windows default ACLs, the new code ignores SYSTEM and Administrators group permissions when computing the MASK/CLASS_OBJ permission mask on old ACLs, and it doesn't deny access to SYSTEM and Administrators group based on the value of MASK/CLASS_OBJ when creating the new ACLs. The new code now handles the S_ISGID bit on directories as on Linux: Setting S_ISGID on a directory causes new files and subdirs created within to inherit its group, rather than the primary group of the user who created the file. This only works for files and directories created by Cygwin processes. - basename(3) now comes in two flavors, POSIX and GNU. The POSIX version is the default. You get the GNU version after #define _GNU_SOURCE #include string.h - The maximum number of PTYs has been raised from 64 to 128. Bug Fixes - - Fix potential hang in pseudo ttys when generating ECHO output while the slave is flooding the pty with output. Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00019.html - Fix potential premature SIGHUP in pty code. Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00070.html - Fix a name change from symlink to target name in calls to execvp, system, etc. Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00270.html - Fix internal error in pty -ONLCR handling. Fix timing bug in pty OPOST handling. Addresses: https://cygwin.com/ml/cygwin/2015-02/msg00929.html NOTE: This change introduces a not yet addressed regression. Native Windows tools generating output with Unix LF instead of Windows CRLF line endings will not get OPOST handling. This prominently affects icacls. - Avoid creating passwd and group records from fully qualified Windows account names (domain\name, name@domain). Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00528.html - Avoid potential crash at startup or in getgroups(2). Addresses: https://cygwin.com/ml/cygwin/2015-04/msg00010.html - Fix UTF-16 surrogate handling in wctomb and friends. Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00452.html To install 32-bit Cygwin use https://cygwin.com/setup-x86.exe To install 64 bit Cygwin use
[ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-0.7
Hi Cygwin friends and users, New 2.0.0-0.7 test release: - Improved setfacl tool. It now handles mask recomputation just like the Linux tool. -d option renamed to -x (but -d is still accepted for backward compat). New -n,--no-mask and --mask options. - Improved getfacl tool. It now prints effective permissions just like the Linux tool. --noname optioned renamed to --numeric (but --noname is still accepted for backward compat). New -c,--omit-header, -e,--all-effective, and -E,--no-effective options. - Fix a bug in account handling. In certain scenarios a PSID pointer was pointing into a buffer space which got moved away due to a call to realloc. This bug seems to be the problem of the reported Permission denied problems when trying to chmod. - Add a missing check for uid/gid values which can't be resolved to Windows accounts. chmod now returns EINVAL now in this case, just as the pre 2.0 DLLs. - I introduced a change in chmod behaviour which is not exactly in line with POSIX 1003.1e draft 17: If you have a non-trivial ACL with secondary accounts and thus a mask value, chmod is supposed to change only the mask, not the permissions of the primary group. However, if the primary group has few permissions to begin with, the result is really surprising. ls -l would, e.g., show read/write perms for the group, but the group might still have only read perms. Personally I find this chmod behaviour really, really bad, so I took the liberty to change it in a way which gives a much less surprising result: If you call chmod on a non-trivial ACL, the group permissions will be used for the primary group and the mask. I hope that this is working out ok. - Fix a bug when reading DENAY ACEs of accounts with multiple entries (typically in the owner == group case). - Fix how group perms are merged into owner perms in owner == group case. - Fix a bug in SID handling which may result in broken SID info in passwd/group entries of unknown accounts. Other than that... The important change in this release is the POSIX permission handling change, a rewrite of the underlying routines reading and creating Windows ACLs following POSIX permission rules and POSIX ACL creating rules per POSIX 1003.1e draft 17, as on Linux. For a description of POSIX ACLs, see http://linux.die.net/man/5/acl All changes in this release so far: === - New, unified implementation of POSIX permission and ACL handling. The new ACLs now store the POSIX ACL MASK/CLASS_OBJ permission mask, and they allow to inherit the S_ISGID bit. ACL inheritance now really works as desired, in a limited, but theoretically equivalent fashion even for non-Cygwin processes. To accommodate Windows default ACLs, the new code ignores SYSTEM and Administrators group permissions when computing the MASK/CLASS_OBJ permission mask on old ACLs, and it doesn't deny access to SYSTEM and Administrators group based on the value of MASK/CLASS_OBJ when creating the new ACLs. The new code now handles the S_ISGID bit on directories as on Linux: Setting S_ISGID on a directory causes new files and subdirs created within to inherit its group, rather than the primary group of the user who created the file. This only works for files and directories created by Cygwin processes. - basename(3) now comes in two flavors, POSIX and GNU. The POSIX version is the default. You get the GNU version after #define _GNU_SOURCE #include string.h - The maximum number of PTYs has been raised from 64 to 128. Bug Fixes - - Fix potential hang in pseudo ttys when generating ECHO output while the slave is flooding the pty with output. Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00019.html - Fix potential premature SIGHUP in pty code. Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00070.html - Fix a name change from symlink to target name in calls to execvp, system, etc. Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00270.html - Fix internal error in pty -ONLCR handling. Fix timing bug in pty OPOST handling. Addresses: https://cygwin.com/ml/cygwin/2015-02/msg00929.html NOTE: This change introduces a not yet addressed regression. Native Windows tools generating output with Unix LF instead of Windows CRLF line endings will not get OPOST handling. This prominently affects icacls. - Avoid creating passwd and group records from fully qualified Windows account names (domain\name, name@domain). Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00528.html - Avoid potential crash at startup or in getgroups(2). Addresses: https://cygwin.com/ml/cygwin/2015-04/msg00010.html - Fix UTF-16 surrogate handling in wctomb and friends. Addresses: https://cygwin.com/ml/cygwin/2015-03/msg00452.html To install 32-bit Cygwin use https://cygwin.com/setup-x86.exe To install 64 bit Cygwin use
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On Fri, Apr 17, 2015 at 6:07 AM, Corinna Vinschen corinna-cyg...@cygwin.com wrote: On Apr 17 10:16, Corinna Vinschen wrote: On Apr 16 12:53, Bryan Berns wrote: On Thu, Apr 16, 2015 at 10:17 AM, Jim Reisert AD1C jjreis...@alum.mit.edu wrote: I am unable to start Cywin/X X-server 1.17.1 with this version. Previous releases of 2.0.0.x were OK. I had to revert to 1.7.35-1 for the time being. Other than updating to 2.0.0.5, I also installed the April 2015 Patch Tuesday updates from Microsoft. I don't know if the two are related. Windows 7 Home Premium, 64-bit Exception: STATUS_ACCESS_VIOLATION at eip=77C50F8A eax= ebx=612D67B0 ecx=1000 edx=612D2648 esi= edi=0028C790 ebp=0028C608 esp=0028C604 program=C:\Cygwin\bin\XWin.exe, pid 1660, thread main cs=0023 ds=002B es=002B fs=0053 gs=002B ss=002B Stack trace: Frame Function Args 0028C608 77C50F8A (, 612D2648, , 612D67B0) 0028C738 610CDA1F (43FF, , , 80012428) 0028C7B8 61047198 (, 72483F24, 75604227, 0254) 0028C7F8 610F629D (0001, , , 75623912) -- Jim Reisert AD1C, jjreis...@alum.mit.edu, http://www.ad1c.us I've actually had problem building Cygwin (1.7.35 or 2.x source) on Cygwin 2.x using Windows. The make errors out with a permissions denied when setting permissions (chmod +x) on config.status in builddir/etc. I was able to produce it on three different, freshly built copies of Windows (no BLODA at all). Only had Cygwin plus the build essentials (gcc, ming, xmlto, diffutils, textinfo, cocom) installed. Only tried with 32-bit. The super wacky thing is that if I rearrange the unrelated contents of the make file that fuels the whole process, the problem goes away. If I downgrade the core back to 1.7.35, the problem goes away. It almost seems like there might an uninitialized memory issue in the code path executed in chmod(). I apologize for the lack of debugging on my part, but I just wanted to throw this out there in case a) it could be related to this issue or b) anyone else has seen this. This is more in line with what Ismail reported in https://cygwin.com/ml/cygwin/2015-04/msg00365.html And I finally managed to reproduce it. It works on the command line and only fails during `make' for me. Investigating now... I think I found the culprit. I'll uploade a -0.7 test release in the next hour or so. Thanks for testing, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat On my way to work now and probably will be offline most of the day, but just wanted to let you know that I did a quick rip of the DLL file from cygwin-2.0.0-0.7.tar.xz on the sourceware.org mirror and make / chmod seems happier now. Nice! :-) -- 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 hangs up if several keys are typed during outputting a lot of texts.
Hi Takashi,
On Apr 17 20:27, Takashi Yano wrote:
Hi Corinna,
On Thu, 16 Apr 2015 11:05:33 +0200
Corinna Vinschen corinna-cyg...@cygwin.com wrote:
Ok, but... this is a really big patch and it complicates the pty code
even more. Is there really no other option as far as the TCSADRAIN
problem is concerned?
What strikes me as weird is that neither fhandler_pty_slave::tcsetattr
nor fhandler_pty_master::tcsetattr give a damn for the optional_actions
parameter. They simply overwrite the tc settings. So I'm wondering,
wouldn't it be possible to add code to the tcsetattr implementation
instead, so that TCSADRAIN/TCSAFLUSH are honored and than only have one
place for OPOST handling?
I also think the patch was a big deal. However, I did not have
any other good idea.
Anyway, I have worked out another solution. Please find a patch
attached.
What do you think of this one?
Looks better to me. However:
@@ -868,6 +980,9 @@ fhandler_pty_slave::tcgetattr (struct termios *t)
int
fhandler_pty_slave::tcsetattr (int, const struct termios *t)
{
+ DWORD n;
+ while (::bytes_available (n, from_slave) n)
+cygwait (10);
acquire_output_mutex (INFINITE);
get_ttyp ()-ti = *t;
release_output_mutex ();
Shouldn't this loop be skipped in TCSANOW mode?
OPOST code has been now completely moved back to master side as
with original implementation.
With this patch, tcsetattr() waits until master reads all data
in pipe before new attributes are applied to preserve the order
between write() and tcsetattr().
However, there is a potential risk in which tcsetattr() can be
blocked if master stops reading pipe, even though I can not imagine
such a likely situation.
Yeah, but it is a busy wait. Hmm. Also, on second thought, the above
loop checks for bytes_available every time and changes n. So, if
another slave writes lots of data, wouldn't the slave calling tcsetattr
starve?
IIUC, what you'd really like to know is something else. It's not about
having n 0 bytes in the pipe, but on calling tcsetattr, you'd like to
know how much bytes are in the pipe at this very moment, and then you'd
overwrite the termios info as soon as these n bytes are written. That
sounds pretty different to me.
It would be cool if the slave-side tcsetattr only transmits the
optional_actions and the new termios content to the master, and the
master keeps it stashed away together with the number of bytes in the
pipe right now. Then in, say, process_slave_output, it checks if the
precondition is fulfilled and only then overwrites its termios
structure.
Off the top of my head I'm not sure how feasible this is. One way to do
that *may* be to send the info in the normal write pipe to the master.
What we need then would be a method to identify such a tcsetattr packet
in the input stream.
Other ideas?
Thanks,
Corinna
--
Corinna Vinschen Please, send mails regarding Cygwin to
Cygwin Maintainer cygwin AT cygwin DOT com
Red Hat
pgpIA8ijrlWUs.pgp
Description: PGP signature
Re: [TESTERS needed] New POSIX permission handling
Hi, Corinna Vinschen-2 wrote On Apr 17 09:30, Corinna Vinschen wrote: I think I found the culprit. I'll uploade a -0.7 test release in the next hour or so. So far all my tests work, thanks a bunch! -- View this message in context: http://cygwin.1069669.n5.nabble.com/TESTERS-needed-New-POSIX-permission-handling-tp117406p117666.html Sent from the Cygwin list mailing list archive at Nabble.com. -- 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
[newlib-cygwin] Support acl(2) method for reading pty ACLs, fix pty chown
https://sourceware.org/git/gitweb.cgi?p=newlib-cygwin.git;h=f63dffb818f9856a43ed6cfb3395d882b21d94b8
commit f63dffb818f9856a43ed6cfb3395d882b21d94b8
Author: Corinna Vinschen cori...@vinschen.de
Date: Fri Apr 17 19:54:59 2015 +0200
Support acl(2) method for reading pty ACLs, fix pty chown
* fhandler.h (fhandler_pty_slave::facl): Add prototype.
* fhandler_tty.cc (fhandler_pty_slave::facl): New method.
(fhandler_pty_slave::fchown): Fix uid/gid handling.
* sec_acl.cc (set_posix_access): Drop superfluous class_idx
variable.
Simplify and move around code in a few places. To improve ACL
readability, add r/w permissions to Admins ACE appended to pty ACL.
Add comment to explain Windows ACE Mask filtering being in the way
of
creating a real CLASS_OBJ.
(get_posix_access): Fake CLASS_OBJ for ptys. Explain why.
* security.cc (get_object_attribute): Add S_IFCHR flag to attributes
when calling get_posix_access.
Signed-off-by: Corinna Vinschen cori...@vinschen.de
Diff:
---
winsup/cygwin/ChangeLog | 14 ++
winsup/cygwin/fhandler.h | 1 +
winsup/cygwin/fhandler_tty.cc | 64 +--
winsup/cygwin/sec_acl.cc | 56 +++--
winsup/cygwin/security.cc | 9 --
5 files changed, 120 insertions(+), 24 deletions(-)
diff --git a/winsup/cygwin/ChangeLog b/winsup/cygwin/ChangeLog
index fb66f39..04bd520 100644
--- a/winsup/cygwin/ChangeLog
+++ b/winsup/cygwin/ChangeLog
@@ -1,5 +1,19 @@
2015-04-17 Corinna Vinschen cori...@vinschen.de
+ * fhandler.h (fhandler_pty_slave::facl): Add prototype.
+ * fhandler_tty.cc (fhandler_pty_slave::facl): New method.
+ (fhandler_pty_slave::fchown): Fix uid/gid handling.
+ * sec_acl.cc (set_posix_access): Drop superfluous class_idx variable.
+ Simplify and move around code in a few places. To improve ACL
+ readability, add r/w permissions to Admins ACE appended to pty ACL.
+ Add comment to explain Windows ACE Mask filtering being in the way of
+ creating a real CLASS_OBJ.
+ (get_posix_access): Fake CLASS_OBJ for ptys. Explain why.
+ * security.cc (get_object_attribute): Add S_IFCHR flag to attributes
+ when calling get_posix_access.
+
+2015-04-17 Corinna Vinschen cori...@vinschen.de
+
* uinfo.cc (pwdgrp::fetch_account_from_windows): Always revert SID
subauth count after checking for known domain.
diff --git a/winsup/cygwin/fhandler.h b/winsup/cygwin/fhandler.h
index 694c23b..05269e5 100644
--- a/winsup/cygwin/fhandler.h
+++ b/winsup/cygwin/fhandler.h
@@ -1545,6 +1545,7 @@ class fhandler_pty_slave: public fhandler_pty_common
select_record *select_read (select_stuff *);
virtual char const *ttyname () { return pc.dev.name; }
int __reg2 fstat (struct stat *buf);
+ int __reg3 facl (int, int, struct acl *);
int __reg1 fchmod (mode_t mode);
int __reg2 fchown (uid_t uid, gid_t gid);
diff --git a/winsup/cygwin/fhandler_tty.cc b/winsup/cygwin/fhandler_tty.cc
index ccb76d9..d243d51 100644
--- a/winsup/cygwin/fhandler_tty.cc
+++ b/winsup/cygwin/fhandler_tty.cc
@@ -12,6 +12,7 @@ details. */
#include winsup.h
#include stdlib.h
#include sys/param.h
+#include sys/acl.h
#include cygwin/kd.h
#include cygerrno.h
#include security.h
@@ -1018,6 +1019,62 @@ fhandler_pty_slave::fstat (struct stat *st)
return 0;
}
+int __reg3
+fhandler_pty_slave::facl (int cmd, int nentries, aclent_t *aclbufp)
+{
+ int res = -1;
+ bool to_close = false;
+ security_descriptor sd;
+ mode_t attr = S_IFCHR;
+
+ switch (cmd)
+{
+ case SETACL:
+ if (!aclsort32 (nentries, 0, aclbufp))
+ set_errno (ENOTSUP);
+ break;
+ case GETACL:
+ if (!aclbufp)
+ {
+ set_errno (EFAULT);
+ break;
+ }
+ /*FALLTHRU*/
+ case GETACLCNT:
+ if (!input_available_event)
+ {
+ char buf[MAX_PATH];
+ shared_name (buf, INPUT_AVAILABLE_EVENT, get_minor ());
+ input_available_event = OpenEvent (READ_CONTROL, TRUE, buf);
+ if (input_available_event)
+ to_close = true;
+ }
+ if (!input_available_event
+ || get_object_sd (input_available_event, sd))
+ {
+ res = get_posix_access (NULL, attr, NULL, NULL, aclbufp, nentries);
+ if (aclbufp res == MIN_ACL_ENTRIES)
+ {
+ aclbufp[0].a_perm = S_IROTH | S_IWOTH;
+ aclbufp[0].a_id = 18;
+ aclbufp[1].a_id = 544;
+ }
+ break;
+ }
+ if (cmd == GETACL)
+ res = get_posix_access (sd, attr, NULL, NULL, aclbufp, nentries);
+ else
+ res = get_posix_access (sd, attr, NULL, NULL, NULL, 0);
+ break;
+ default:
+ set_errno (EINVAL);
+ break;
+
Re: [TESTERS needed] New POSIX permission handling
On Apr 17 08:17, Ismail Donmez wrote: Hi, Corinna Vinschen-2 wrote On Apr 17 09:30, Corinna Vinschen wrote: I think I found the culprit. I'll uploade a -0.7 test release in the next hour or so. So far all my tests work, thanks a bunch! Thanks for your feedback! Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpN_BzNCPHoo.pgp Description: PGP signature
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-5
On Apr 17 06:17, Ian Lambert wrote: Try the -0.6 I'm just uploading. This should work again since it checks the PSIDs for NULL before calling this crashing OS function. Indeed, it works again. Thanks! And ssh works too. :) It still gives this error message, but it was only noticed because it was the last line of output: winMultiWindowXMsgProcErrorHandler - ERROR: BadMatch (invalid parameter attributes) This is some XWin issue. I doubt that it has something to do with the ACL handling. Thanks, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpPvzcYT8SoN.pgp Description: PGP signature
Re: Running tasklist /m in cygwin hangs
On 4/17/2015 5:23 PM, Saurabh T wrote: I can confirm it still hangs for me with 1.7.35. I am using the 32 bit (i686) Cygwin which is the only difference from Andrey's. on $ uname -vrsm CYGWIN_NT-6.1-WOW 1.7.35(0.287/5/3) 2015-03-04 12:07 i686 It is only a bit slow (both mintty and console) $ time tasklist /m normaliz.DLL Image Name PID Modules = ZenworksWindowsService.ex 1424 normaliz.dll AppleMobileDeviceService. 1616 normaliz.dll explorer.exe 2520 normaliz.dll PanGPS.exe3276 normaliz.dll PanGPA.exe3420 normaliz.dll iTunesHelper.exe 3544 normaliz.dll Smc.exe 5008 normaliz.dll real1m25.565s user0m0.015s sys 0m0.046s It seems some type of timeout in the interaction of this windows utility and cygwin Regards Marco -- 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: TEST RELEASE: Cygwin 2.0.0-0.7
On Fri, Apr 17, 2015 at 7:46 AM, Jim Reisert AD1C wrote: Is there a way to change my default home directory without using /etc/passwd, other than creating a symbolic link inside of /home? You can modify /etc/fstab like this c:/home /home . acl Example http://github.com/svnpenn/dotfiles/blob/24302d29ef7158df1aa3a584131c810/fstab#L2 -- 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: dig utility in bind-utils-9.9.7-1 returns no output
Yaakov, Brian, Andrey, Marco, and Mark - I updated Cygwin this morning, and now: dig works! I have no idea why! cksum /usr/bin/dig /bin/dig reports the same checksums as before, and cygcheck dig produces the same output as before. Baffling. Thanks to all of you for the suggestions. On Wed, Apr 15, 2015 at 4:19 PM, Keith Christian keith1christ...@gmail.com wrote: Thanks Mark, but I tried both of those early in my attempts without success. Good suggestion, though. On Wed, Apr 15, 2015 at 2:53 PM, Mark Hansen m...@winfirst.com wrote: On 4/15/2015 12:02 PM, Keith Christian wrote: cygcheck attachment is too large according to sourceware. What specific parts of cygcheck are needed to help troubleshoot this? Keith On Wed, Apr 15, 2015 at 12:55 PM, Keith Christian keith1christ...@gmail.com wrote: Both /usr/bin/dig and /bin/dig run the same program: cksum /usr/bin/dig /bin/dig 680300229 2202141 /usr/bin/dig 680300229 2202141 /bin/dig Attached cygcheck_s_r_v_kvc.txt. I'm sorry - I could be completely off-base here. What I meant was can you try running '/usr/bin/dig' (providing the full path to the application) rather than by just running 'dig' just in case there is some other 'dig' on your path (or shell aliases) which might be running rather than the actual /usr/bin/dig command. In previous messages, your examples were running 'dig' (without the full path) so it wasn't clear which dig you were actually getting. Good luck. On Tue, Apr 14, 2015 at 3:53 PM, Mark Hansenm...@winfirst.com wrote: Are you sure that you're getting the correct dig executable? Is it possible that something else in your path or alias list is getting hit? Perhaps you can try /usr/bin/dig just to be sure? -- 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 -- 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: TEST RELEASE: Cygwin 2.0.0-0.7
https://cygwin.com/cygwin-ug-net/ntsec.html#ntsec-mapping-nsswitch-home Very helpful, Corinna! Working as advertised. - Jim -- Jim Reisert AD1C, jjreis...@alum.mit.edu, http://www.ad1c.us -- 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
NTSEC db_home
I'm having a little trouble with the db_home: setting at work (corporate network, not using /etc/passwd or /etc/group). Regarding: https://cygwin.com/cygwin-ug-net/ntsec.html#ntsec-mapping-nsswitch-home I tried db_home: /%H and although my MINTTY session started in the correct directory, ssh-keygen wanted to use the directory defined by HOMESHARE instead, which points somewhere else (a local Unix server). Note that both HOMEDRIVE (+HOMEPATH) and HOMESHARE ultimately point to the same place: DOS view of the world: c:\set | grep HOME HOME=C:\Home HOMEDRIVE=P: HOMEPATH=\ HOMESHARE=\\corp.ad.broadcom.com\DENA\home\reisert (mintty window with bash) [LTDENA-REISERT:~] $ ssh-keygen Generating public/private rsa key pair. Enter file in which to save the key (//corp.ad.broadcom.com/DENA/home/reisert/.ssh/id_rsa): Cygwin view of the world: [LTDENA-REISERT:~] $ set | grep -i home CDPATH=.:/cygdrive/c/Home:/cygdrive/c/Home/dx4win:/cygdrive/c HISTFILE=/cygdrive/c/Home/.bash_history HOME=/cygdrive/c/Home HOMEDRIVE=P: HOMEPATH='\' HOMESHARE='\\corp.ad.broadcom.com\DENA\home\reisert' PWD=/cygdrive/c/Home I can't explain what's going on. I changed db_home to: db_home: /cygdrive/C/Home and it's working as expected. -- Jim Reisert AD1C, jjreis...@alum.mit.edu, http://www.ad1c.us -- 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: R help leaves out lines of text
Marco Atzeri marco.atzeri at gmail.com writes:
I am ramping up on the R statistical analysis environment. I find
that the help leaves out entire lines of text. The pager is
/usr/lib/R/bin/pager. I changed it to /bin/less, but I see the
same symptom. The problem shows up both in xterm and mintty. The
computer is in a locked down environment, so a straight update of
cygwin packages is not an option. From the R forum, the problem
might be due to improper handling of line endings by the R port to
cygwin.
Is there anything further I can try to circumvent the problem?
My version info is:
64-bit Cygwin DLL version 1.7.28
R version 3.0.1-1
Are the input files Unix style or DOS style ?
Can you give an example to replicate your finding ?
I can that test than if the last R-3.1.3-1 package behaves correctly
or not. R-3.0.1-1 package is almost 2 years old.
I can't actually find the help files; I suspect that they are not just
plain text. I tried looking for html versions of the help pages by
ferruting through the R.home() subtree. Haven't found them so far.
There are package pages in subdirectories package/html/00Index.html,
but they just contain links to html files that don't reside in my
R.home() subtree. There are also subdirectories package/help, but
they contain pages that I don't recognize (*.rds, *.rdb, *.rdx).
As an example, I've attached the output from ?help at the very end
below. My current workaround is to use the following in my
~/.Rprofile, which forces the help to a browser:
options(browser=cygstart)
options(help_type=html)
Here's the bad result from help(help,help_type=text):
help package:utils R Documentation
multiple blank lines...
Documentation
multiple blank lines...
Description:
multiple blank lines...
Usage:
multiple blank lines...
help(topic, package = NULL, lib.loc = NULL,
verbose = getOption(verbose),
try.all.packages = getOption(help.try.all.packages),
help_type = getOption(help_type))
multiple blank lines...
Arguments:
multiple blank lines...
lines 1-23- (less prompt)
multiple blank lines...
Details:
multiple blank lines...
The following types of help are available:
multiple blank lines...
Plain text help
multiple blank lines...
If a package is specified, (text or, in interactive use only, HTML)
information on the package, including hints/links to suitable
help topics.
multiple blank lines...
lines 24-46- (less prompt)
multiple blank lines...
Offline help:
multiple blank lines...
Note:
multiple blank lines...
References:
multiple blank lines...
Becker, R. A., Chambers, J. M. and Wilks, A. R. (1988) _The New S
Language_. Wadsworth Brooks/Cole.
multiple blank lines...
See Also:
multiple blank lines...
Examples:
multiple blank lines...
help()
help(help) # the same
multiple blank lines...
lines 47-69- (less prompt)
help(lapply)
multiple blank lines...
help(for) # or ?for, but quotes/backticks are needed
multiple blank lines...
try({# requires working TeX installation:
help(dgamma, help_type = pdf)
## - nicely formatted pdf -- including math formula -- for help
(dgamma):
system2(getOption(pdfviewer), dgamma.pdf, wait = FALSE)
})
multiple blank lines...
help(package = splines) # get help even when package is not loaded
multiple blank lines...
topi - women
help(topi)
multiple blank lines...
try(help(bs, try.all.packages = FALSE)) # reports not found (an
error)
help(bs, try.all.packages = TRUE) # reports can be found
# in package 'splines'
multiple blank lines...
## For programmatic use:
topic - family; pkg_ref - stats
help((topic), (pkg_ref))
multiple blank lines...
lines 70-92- (less prompt)
multiple blank lines...
help(for) # or ?for, but quotes/backticks are needed
multiple blank lines...
try({# requires working TeX installation:
help(dgamma, help_type = pdf)
## - nicely formatted pdf -- including math formula -- for help
(dgamma):
system2(getOption(pdfviewer), dgamma.pdf, wait = FALSE)
})
multiple blank lines...
help(package = splines) # get help even when package is not loaded
multiple blank lines...
topi - women
help(topi)
multiple blank lines...
try(help(bs, try.all.packages = FALSE)) # reports not found (an
error)
help(bs, try.all.packages = TRUE) # reports can be found
# in package 'splines'
multiple blank lines...
## For programmatic use:
topic -
Re: [ANNOUNCEMENT] TEST RELEASE: Cygwin 2.0.0-0.7
Corinna Vinschen writes: New 2.0.0-0.7 test release: - Improved setfacl tool. It now handles mask recomputation just like the Linux tool. -d option renamed to -x (but -d is still accepted for backward compat). New -n,--no-mask and --mask options. setfacl -b -k still errors out instead of removing both the default and extended ACL entries. The important change in this release is the POSIX permission handling change, a rewrite of the underlying routines reading and creating Windows ACLs following POSIX permission rules and POSIX ACL creating rules per POSIX 1003.1e draft 17, as on Linux. I seem to have found another fly in that ointment (or rather cygport did find it for me…): While packaging a find usr/ -type f -executable would find newly created info files that ls and getfacl agree are not executable: -rw---+ 1 ASSI Kein 48880 5. Apr 2014 ucl.log # file: ucl.log # owner: ASSI # group: Kein user::rw- group::--- group:SYSTEM:rwx#effective:--- group:Administratoren:rwx #effective:--- mask:--- other:--- It seems that some of the code doesn't take the masking bits into account just yet. Here's the relevant portion of an strace on a different file (I had already deleted the ACL on the original ones): 215 43687 [main] find 2808 normalize_posix_path: src /mnt/share/maint/ucl.log 23 43710 [main] find 2808 normalize_posix_path: /mnt/share/maint/ucl.log = normalize_posix_path (/mnt/share/maint/ucl.log) 35 43745 [main] find 2808 mount_info::conv_to_win32_path: conv_to_win32_path (/mnt/share/maint/ucl.log) 28 43773 [main] find 2808 set_flags: flags: binary (0x2) 27 43800 [main] find 2808 mount_info::conv_to_win32_path: src_path /mnt/share/maint/ucl.log, dst D:\Freeware\CygShare\maint\ucl.log, flags 0x2, rc 0 63 43863 [main] find 2808 symlink_info::check: 0x0 = NtCreateFile (\??\D:\Freeware\CygShare\maint\ucl.log) 43 43906 [main] find 2808 symlink_info::check: not a symlink 57 43963 [main] find 2808 symlink_info::check: 0 = symlink.check(D:\Freeware\CygShare\maint\ucl.log, 0x23B600) (0x42) 24 43987 [main] find 2808 path_conv::check: this-path(D:\Freeware\CygShare\maint\ucl.log), has_acls(1) 32 44019 [main] find 2808 build_fh_pc: fh 0x1802FEE38, dev 00C3 22 44041 [main] find 2808 stat_worker: (\??\D:\Freeware\CygShare\maint\ucl.log, 0x60003B6E0, 0x1802FEE38), file_attributes 32 125 44166 [main] find 2808 fhandler_base::fstat_helper: 0 = fstat (\??\D:\Freeware\CygShare\maint\ucl.log, 0x60003B6E0) st_size=48880, st_mode=0100600, st_ino=281474977000743st_atim=533FAD65.25D44958 st_ctim=5531679B.D5E5E90 st_mtim=533FADD7.6CDF1E4 st_birthtim=533FAD65.25D44958 27 44193 [main] find 2808 stat_worker: 0 = (\??\D:\Freeware\CygShare\maint\ucl.log,0x60003B6E0) 226 44419 [main] find 2808 normalize_posix_path: src /mnt/share/maint/ucl.log 23 2 [main] find 2808 normalize_posix_path: /mnt/share/maint/ucl.log = normalize_posix_path (/mnt/share/maint/ucl.log) 21 44463 [main] find 2808 mount_info::conv_to_win32_path: conv_to_win32_path (/mnt/share/maint/ucl.log) 25 44488 [main] find 2808 set_flags: flags: binary (0x2) 19 44507 [main] find 2808 mount_info::conv_to_win32_path: src_path /mnt/share/maint/ucl.log, dst D:\Freeware\CygShare\maint\ucl.log, flags 0x2, rc 0 86 44593 [main] find 2808 symlink_info::check: 0x0 = NtCreateFile (\??\D:\Freeware\CygShare\maint\ucl.log) 34 44627 [main] find 2808 symlink_info::check: not a symlink 21 44648 [main] find 2808 symlink_info::check: 0 = symlink.check(D:\Freeware\CygShare\maint\ucl.log, 0x23B530) (0x42) 21 44669 [main] find 2808 path_conv::check: this-path(D:\Freeware\CygShare\maint\ucl.log), has_acls(1) 38 44707 [main] find 2808 build_fh_pc: fh 0x1802FEE38, dev 00C3 110 44817 [main] find 2808 check_file_access: flags 0x1, ret 0 47 44864 [main] find 2808 fhandler_base::fhaccess: returning 0 44 44908 [main] find 2808 faccessat: returning 0 With the ACL removed, the trace looks like this instead: 277 46985 [main] find 3992 normalize_posix_path: src /mnt/share/maint/ucl.log 29 47014 [main] find 3992 normalize_posix_path: /mnt/share/maint/ucl.log = normalize_posix_path (/mnt/share/maint/ucl.log) 39 47053 [main] find 3992 mount_info::conv_to_win32_path: conv_to_win32_path (/mnt/share/maint/ucl.log) 20 47073 [main] find 3992 set_flags: flags: binary (0x2) 18 47091 [main] find 3992 mount_info::conv_to_win32_path: src_path /mnt/share/maint/ucl.log, dst D:\Freeware\CygShare\maint\ucl.log, flags 0x2, rc 0 58 47149 [main] find 3992 symlink_info::check: 0x0 = NtCreateFile (\??\D:\Freeware\CygShare\maint\ucl.log) 51 47200 [main] find 3992 symlink_info::check: not a symlink 20 47220 [main] find 3992 symlink_info::check: 0 = symlink.check(D:\Freeware\CygShare\maint\ucl.log, 0x23B600) (0x42) 23 47243 [main] find 3992
readline behaves differently in R than in bash
I set editing-mode vi in ~/.inputrc. In bash, it responds as expected to (say) dt,, which deletes characters upto the next comma. It does not do this in R; instead, that series of keystrokes simply causes the next two keystrokes to be consumed without any effect. I guess it would be wrong to assume that R uses the same readline implementation as bash? Is there anything I can set to get the expected behaviour? 64-bit Cygwin DLL version 1.7.28 R version 3.0.1-1 -- 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: [ANNOUNCEMENT] Updated: lilypond-2.19.18-2 [GOLDSTAR]
On Fri, 2015-04-17 at 06:13 +0200, Marco Atzeri wrote: Versions 2.19.18-2 of lilypond lilypond-doc for cygwin are now available: CYGWIN CHANGES Both architecture are now available using latest development (but more cygwin 32bit robust) release Thanks for getting this working again. Andrew, could you do the honours? -- 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: less color option
Steven Penny sent the following at Friday, April 17, 2015 5:08 PM man less Reveals this -Dxcolor or --color=xcolor [MS-DOS only] Sets the color of the text displayed. x is a sin? gle character which selects the type of text whose color is being set: n=normal, s=standout, d=bold, u=underlined, k=blink. color is a pair of numbers separated by a period. The first number selects the foreground color and the second selects the background color of the text. A single number N is the same as N.M, where M is the normal background color. So I tried it and get this result $ less --color There is no color option (less --help for help) I just want a way to change the colors on man pages. If this is not the right way please advise. Just a guess: The man page says [MS-DOS only]. Cygwin is not MS-DOS so I wouldn't expect this to work with less compiled for cygwin. If you were using the MS-DOS version, you would still need to specify xcolor. Hope that helps, - Barry Disclaimer: Statements made herein are not made on behalf of NIAID.
MATE category ?
Yaakov, there are several device with MATE as category. Are we going over the list defined in: https://cygwin.com/setup.html Regards Marco
less color option
man less Reveals this -Dxcolor or --color=xcolor [MS-DOS only] Sets the color of the text displayed. x is a sin‐ gle character which selects the type of text whose color is being set: n=normal, s=standout, d=bold, u=underlined, k=blink. color is a pair of numbers separated by a period. The first number selects the foreground color and the second selects the background color of the text. A single number N is the same as N.M, where M is the normal background color. So I tried it and get this result $ less --color There is no color option (less --help for help) I just want a way to change the colors on man pages. If this is not the right way please advise. -- 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
Updated: qt4-4.8.6-6, qt4-webkit-4.10.4-1
The following packages have been updated for the Cygwin distribution: * libQt3Support4-4.8.6-6 * libQt3Support4-devel-4.8.6-6 * libQtCore4-4.8.6-6 * libQtCore4-devel-4.8.6-6 * libQtDBus4-4.8.6-6 * libQtDBus4-devel-4.8.6-6 * libQtDeclarative4-4.8.6-6 * libQtDeclarative4-devel-4.8.6-6 * libQtDesigner4-4.8.6-6 * libQtDesigner4-devel-4.8.6-6 * libQtGui4-4.8.6-6 * libQtGui4-devel-4.8.6-6 * libQtHelp4-4.8.6-6 * libQtHelp4-devel-4.8.6-6 * libQtMultimedia4-4.8.6-6 * libQtMultimedia4-devel-4.8.6-6 * libQtNetwork4-4.8.6-6 * libQtNetwork4-devel-4.8.6-6 * libQtOpenGL4-4.8.6-6 * libQtOpenGL4-devel-4.8.6-6 * libQtScript4-4.8.6-6 * libQtScript4-devel-4.8.6-6 * libQtSql4-4.8.6-6 * libQtSql4-devel-4.8.6-6 * libQtSvg4-4.8.6-6 * libQtSvg4-devel-4.8.6-6 * libQtTest4-4.8.6-6 * libQtTest4-devel-4.8.6-6 * libQtWebKit4-4.10.4-1 * libQtWebKit4-devel-4.10.4-1 * libQtXml4-4.8.6-6 * libQtXml4-devel-4.8.6-6 * libQtXmlPatterns4-4.8.6-6 * libQtXmlPatterns4-devel-4.8.6-6 * qt4-devel-tools-4.8.6-6 * qt4-designer-plugin-webkit-4.8.6-6 * qt4-doc-4.8.6-6 * qt4-qtconfig-4.8.6-6 Qt is a cross-platform application framework for desktop and embedde development. Qt enables programmers to create advanced GUI applications once and deploy them to Windows, Mac OS X and Linux without rewriting the source code. This release drops usage of the older QtWebKit bundled in the upstream qt4 sources in favour of a newer standalone version from KDE. -- Yaakov
[ANNOUNCEMENT] Updated: qt4-4.8.6-6, qt4-webkit-4.10.4-1
The following packages have been updated for the Cygwin distribution: * libQt3Support4-4.8.6-6 * libQt3Support4-devel-4.8.6-6 * libQtCore4-4.8.6-6 * libQtCore4-devel-4.8.6-6 * libQtDBus4-4.8.6-6 * libQtDBus4-devel-4.8.6-6 * libQtDeclarative4-4.8.6-6 * libQtDeclarative4-devel-4.8.6-6 * libQtDesigner4-4.8.6-6 * libQtDesigner4-devel-4.8.6-6 * libQtGui4-4.8.6-6 * libQtGui4-devel-4.8.6-6 * libQtHelp4-4.8.6-6 * libQtHelp4-devel-4.8.6-6 * libQtMultimedia4-4.8.6-6 * libQtMultimedia4-devel-4.8.6-6 * libQtNetwork4-4.8.6-6 * libQtNetwork4-devel-4.8.6-6 * libQtOpenGL4-4.8.6-6 * libQtOpenGL4-devel-4.8.6-6 * libQtScript4-4.8.6-6 * libQtScript4-devel-4.8.6-6 * libQtSql4-4.8.6-6 * libQtSql4-devel-4.8.6-6 * libQtSvg4-4.8.6-6 * libQtSvg4-devel-4.8.6-6 * libQtTest4-4.8.6-6 * libQtTest4-devel-4.8.6-6 * libQtWebKit4-4.10.4-1 * libQtWebKit4-devel-4.10.4-1 * libQtXml4-4.8.6-6 * libQtXml4-devel-4.8.6-6 * libQtXmlPatterns4-4.8.6-6 * libQtXmlPatterns4-devel-4.8.6-6 * qt4-devel-tools-4.8.6-6 * qt4-designer-plugin-webkit-4.8.6-6 * qt4-doc-4.8.6-6 * qt4-qtconfig-4.8.6-6 Qt is a cross-platform application framework for desktop and embedde development. Qt enables programmers to create advanced GUI applications once and deploy them to Windows, Mac OS X and Linux without rewriting the source code. This release drops usage of the older QtWebKit bundled in the upstream qt4 sources in favour of a newer standalone version from KDE. -- 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
[ANNOUNCEMENT] Updated: man-db-2.7.1-1
The following package has been updated in the Cygwin distribution: * man-db-2.7.1-1 man-db is an implementation of the standard Unix documentation system accessed using the man command. It uses an embedded database in place of the traditional flat-text whatis databases. This is an update to the latest upstream release. The default man_db.conf has been changed to support SECTIONs 0p, 1p, and 3p for POSIX documentation. -- 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
Updated: man-db-2.7.1-1
The following package has been updated in the Cygwin distribution: * man-db-2.7.1-1 man-db is an implementation of the standard Unix documentation system accessed using the man command. It uses an embedded database in place of the traditional flat-text whatis databases. This is an update to the latest upstream release. The default man_db.conf has been changed to support SECTIONs 0p, 1p, and 3p for POSIX documentation. -- Yaakov
Re: R help leaves out lines of text
On 4/17/2015 8:57 PM, paul wrote: Marco Atzeri marco.atzeri at gmail.com writes: I am ramping up on the R statistical analysis environment. I find that the help leaves out entire lines of text. The pager is /usr/lib/R/bin/pager. I changed it to /bin/less, but I see the same symptom. The problem shows up both in xterm and mintty. The computer is in a locked down environment, so a straight update of cygwin packages is not an option. From the R forum, the problem might be due to improper handling of line endings by the R port to cygwin. Is there anything further I can try to circumvent the problem? My version info is: 64-bit Cygwin DLL version 1.7.28 R version 3.0.1-1 Are the input files Unix style or DOS style ? Can you give an example to replicate your finding ? I can that test than if the last R-3.1.3-1 package behaves correctly or not. R-3.0.1-1 package is almost 2 years old. I can't actually find the help files; the documentation is available as info files. $ cygcheck -l R |grep info /usr/share/info/R-admin.info.gz /usr/share/info/R-data.info.gz /usr/share/info/R-exts.info-1.gz /usr/share/info/R-exts.info-2.gz /usr/share/info/R-exts.info.gz /usr/share/info/R-FAQ.info.gz /usr/share/info/R-intro.info.gz /usr/share/info/R-ints.info.gz /usr/share/info/R-lang.info.gz For what I can see the help data are also present. The formatting could be better but I have no problem in reading. If you are looking for a HTML browser help, of course it is not configured. Regards Marco -- 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: Running tasklist /m in cygwin hangs
Greetings, Marco Atzeri! I can confirm it still hangs for me with 1.7.35. I am using the 32 bit (i686) Cygwin which is the only difference from Andrey's. on $ uname -vrsm CYGWIN_NT-6.1-WOW 1.7.35(0.287/5/3) 2015-03-04 12:07 i686 It is only a bit slow (both mintty and console) $ time tasklist /m normaliz.DLL Image Name PID Modules = ZenworksWindowsService.ex 1424 normaliz.dll AppleMobileDeviceService. 1616 normaliz.dll explorer.exe 2520 normaliz.dll PanGPS.exe3276 normaliz.dll PanGPA.exe3420 normaliz.dll iTunesHelper.exe 3544 normaliz.dll Smc.exe 5008 normaliz.dll real1m25.565s Oh... wow... real0m0.088s [x].[x] user0m0.015s sys 0m0.046s It seems some type of timeout in the interaction of this windows utility and cygwin Was it native console or pty ? -- With best regards, Andrey Repin Saturday, April 18, 2015 02:02:39 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
[ANNOUNCEMENT] New: qt5-webkit-5.3.2-2, qt5-webkit-examples-5.3.2-1
The following packages have been added to the Cygwin distribution: * libQt5WebKit5-5.3.2-2 * libQt5WebKit-devel-5.3.2-2 * qt5-webkit-doc-5.3.2-2 * qt5-webkit-examples-5.3.2-1 * qt5-webkit-examples-doc-5.3.2-1 These releases add support for the WebKit component to Qt5. -- 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: readline behaves differently in R than in bash
On 4/17/2015 10:28 PM, paul wrote: I set editing-mode vi in ~/.inputrc. In bash, it responds as expected to (say) dt,, which deletes characters upto the next comma. It does not do this in R; instead, that series of keystrokes simply causes the next two keystrokes to be consumed without any effect. I guess it would be wrong to assume that R uses the same readline implementation as bash? Is there anything I can set to get the expected behaviour? 64-bit Cygwin DLL version 1.7.28 R version 3.0.1-1 it works fine for me. $ uname -svr CYGWIN_NT-6.1-WOW 1.7.35(0.287/5/3) 2015-03-04 12:07 $ R --version R version 3.1.3 (2015-03-09) -- Smooth Sidewalk Regards Marco -- 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
New: qt5-webkit-5.3.2-2, qt5-webkit-examples-5.3.2-1
The following packages have been added to the Cygwin distribution: * libQt5WebKit5-5.3.2-2 * libQt5WebKit-devel-5.3.2-2 * qt5-webkit-doc-5.3.2-2 * qt5-webkit-examples-5.3.2-1 * qt5-webkit-examples-doc-5.3.2-1 These releases add support for the WebKit component to Qt5. -- Yaakov
Re: dig utility in bind-utils-9.9.7-1 returns no output
Greetings, Keith Christian! I'm sorry - I could be completely off-base here. What I meant was can you try running '/usr/bin/dig' (providing the full path to the application) rather than by just running 'dig' just in case there is some other 'dig' on your path (or shell aliases) which might be running rather than the actual /usr/bin/dig command. In previous messages, your examples were running 'dig' (without the full path) so it wasn't clear which dig you were actually getting. Thanks Mark, but I tried both of those early in my attempts without success. Good suggestion, though. I updated Cygwin this morning, and now: dig works! I have no idea why! cksum /usr/bin/dig /bin/dig reports the same checksums as before, and cygcheck dig produces the same output as before. Baffling. I've seen similar effect, when Cygwin was going through implementation of some specific POSIX calls, and export tables were bouncing around libs in the snapshots. But your claim that you've had Cygwin updated led the discussion sideways. My bet was you've had Cygwin1.dll locked when updating and the update wasn't completely successful. Thanks to all of you for the suggestions. No problem. However, please don't top-post in future. -- With best regards, Andrey Repin Saturday, April 18, 2015 01:55:23 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: Running tasklist /m in cygwin hangs
On 4/18/2015 1:05 AM, Andrey Repin wrote: Greetings, Marco Atzeri! I can confirm it still hangs for me with 1.7.35. I am using the 32 bit (i686) Cygwin which is the only difference from Andrey's. on $ uname -vrsm CYGWIN_NT-6.1-WOW 1.7.35(0.287/5/3) 2015-03-04 12:07 i686 It is only a bit slow (both mintty and console) $ time tasklist /m normaliz.DLL Image Name PID Modules = ZenworksWindowsService.ex 1424 normaliz.dll AppleMobileDeviceService. 1616 normaliz.dll explorer.exe 2520 normaliz.dll PanGPS.exe3276 normaliz.dll PanGPA.exe3420 normaliz.dll iTunesHelper.exe 3544 normaliz.dll Smc.exe 5008 normaliz.dll real1m25.565s Oh... wow... real0m0.088s [x].[x] user0m0.015s sys 0m0.046s It seems some type of timeout in the interaction of this windows utility and cygwin Was it native console or pty ? on 32bit cygwin on 64bit W7 cmd+bash and mintty+bash have the same long time. As the effect is around 10 sec per line and tasklist.exe is looking which process is using a certain dll, I presume that there is a negative interactions (timeout, double sleep ?). cmd alone or 64 bit cygwin has no delay effect. However on 64bit you can compare /cygdrive/c/windows/SysWOW64/tasklist.exe /m normaliz.dll /cygdrive/c/windows/System32/tasklist.exe /m normaliz.dll timing ;-) Regards Marco -- 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
issues seen in TEST RELEASE: Cygwin 2.0.0-0.7
Hi again, Corinna. I appreciate these recent changes, the more complete Posix ACL support looks beneficial for sharing/syncing files between Cygwin and Linux machines, and for more compatible scripting. But I've noticed a few possibly-concerning items: Item 1: - I introduced a change in chmod behaviour which is not exactly in line with POSIX 1003.1e draft 17: This change looks to me to be somewhat dangerous. I'm concerned it may lead to granting privileges one doesn't expect to have granted. And scripts leveraging this would behave noticeably differently on Cygwin than they do on Linux, leading to extra time to debug/recode. A simplistic example use case: # Create scripts for some app, with group set to give httpd or some # other application login user read and execute privs. For sake of # brevity, here only showing one directory, but reality would likely # be a tree of dirs and files. At least to my sensibilities, having # the primary group be httpd here makes sense in what I'd like to see # in 'ls', and what I'd like 'find' to be able to find most easily, # even tho httpd is intended to have fewer privileges than the # developers group will be given at times on these dirs/files. mkdir appdir chgrp httpd appdir chmod 750 appdir ls -ald appdir getfacl appdir # Give developers write access. setfacl -m g:developers:rwx appdir ls -ald appdir getfacl appdir # Production use period, try to prevent changes. chmod ugo-w appdir ls -ald appdir getfacl appdir # Development period: Reenable changes by local-machine developers - # but did NOT intend to enable writes from the http daemon which may # be up running other applications. On Linux, remains 'group::r-x', # only mask gets changed. chmod g+w appdir ls -ald appdir getfacl appdir --- Item 2: I'm having troubles figuring out how to create a directory that plays well both for providing old-style umask behavior and as a holder of files created by non-Cygwin Windows apps. Cygwin mkdir seems to create some to-be-inherited ACLs, even if the parent directory has had 'setfacl -k' applied (and so it's clear these aren't being inherited down, they seem newly added by mkdir or some library call it makes): CREATOR OWNER:(OI)(CI)(IO)(F) CREATOR GROUP:(OI)(CI)(IO)(RX,W,DC) Everyone:(OI)(CI)(IO)(RX,W,DC) (The actual permissions in these seem to vary based on the umask setting and any inherited ACLs at time of directory creation.) This seems existing behavior in Cygwin, still present in this test drop. What I seem to be seeing is that with these present, this test drop appears to now be treating the directory as having extended ACLs (is that the proper phrasing?), and so now will use these default: ACL entries to create the permissions for a newly-created-by-Cygwin-app file rather than using umask. So even with umask = 0, I now get only 0700 on newly created files in a directory that itself has 0700, rather than 0777 as I get on older Cygwin and on Linux. (What I actually get seems to vary based on the directory-creation-time umask, or whatever ACLs the parent directory itself inherited via this mechanism, as noted in prior paragraph's parenthetical.) I can use 'setfacl -k .' to remove these, but then if I try to write a file into the directory using a non-Cygwin program I end up with a rather strange set of ACLs on it. (I had in past remembered stuff like CURRENT SESSION, but doing a test now using 'echo x file' under cmd I find I get my user(F), cyg_server(F), Administrators(F), and SYSTEM(F) tho none of cyg_server, Administrators, nor SYSTEM are present at all in the ACLs of the parent directory, and my user isn't cyg_server.) I suspect the desire to allow non-Cygwin apps to sanely create files in a Cygwin-created directory is likely the reason that mkdir is creating these ACL entries. Would it maybe make sense for 'setfacl -k' to keep/recreate the same base set of to-be-inherited ACLs that mkdir does, and for the determination of whether to use umask vs. the default: ACLs to ignore these? Maybe for better cosmetics, for getfacl to not show these, nor ls to show these cases only as causing a '+' to appear, so that a newly-created directory with no ACLs inherited from above would appear both visually and behaviorally as not using extended ACLs? I note that chmod doesn't keep these to-be-inherited entries in sync with the directory's mode; they seem stuck at their create-time values unless changed explicitly with icacls. (And am I right that these can't be easily changed with setfacl, since the SIDs involved don't have Cygwin user/group mappings?) This seems true of both this test drop and older Cygwin. Intuitively this feels not-quite-right, but seems a separable concern from the above and isn't a current change. --- Item 3: It seems to create rather messy/redundant ACLs if one tries to use setfacl to adjust the owning user's privileges explicitly by user name. This is also seeming to be interacting with the Item 1 topic,
RE: Running tasklist /m in cygwin hangs
I can confirm it still hangs for me with 1.7.35. I am using the 32 bit (i686) Cygwin which is the only difference from Andrey's. Date: Wed, 15 Apr 2015 22:15:57 +0300 From: anrdae...@yandex.ru To: saur...@hotmail.com; cygwin@cygwin.com Subject: Re: Running tasklist /m in cygwin hangs Greetings, Saurabh T! Running tasklist /m file.dll hangs in Cygwin First, mintty (I assume you did mean that) is not Cygwin itself. even though it works perfectly fine in the cmd window. I hope you did mean native console by that. Is there any reason for this? I am using a somewhat older cygwin (1.7.25) on a Windows 7 box, and do not want to upgrade unless newer cygwin doesnt show the problem. Thank you. $ tasklist /m normaliz.DLL | iconv -f CP866 Имя образа PID Модули = taskhost.exe 3492 normaliz.DLL litestep.exe 2172 normaliz.DLL spideragent.exe 3336 normaliz.DLL Results are the same if running in mintty or native console. No hangs can be seen from my side. CYGWIN_NT-6.1 daemon2 1.7.35(0.287/5/3) 2015-03-04 12:09 x86_64 Cygwin CYGWIN_NT-6.1 daemon2 2.0.0(0.287/5/3) 2015-04-15 17:39 x86_64 Cygwin -- With best regards, Andrey Repin Wednesday, April 15, 2015 22:07:13 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: NTSEC db_home
On Apr 17 10:07, Jim Reisert AD1C wrote: I'm having a little trouble with the db_home: setting at work (corporate network, not using /etc/passwd or /etc/group). Regarding: https://cygwin.com/cygwin-ug-net/ntsec.html#ntsec-mapping-nsswitch-home I tried db_home: /%H and although my MINTTY session started in the correct directory, ssh-keygen wanted to use the directory defined by HOMESHARE instead, which points somewhere else (a local Unix server). Note that both HOMEDRIVE (+HOMEPATH) and HOMESHARE ultimately point to the same place: DOS view of the world: c:\set | grep HOME HOME=C:\Home Bzzz. You seem to misinterpret what %H means. It's the same thing as setting db_home to the windows scheme. Here's the relevant part from the docs: windows The user's home directory is set to the same directory which is used as Windows home directory. This is the homeDirectory AD attribute. For SAM accounts, this is equivalent to the Home folder setting in SAM. If both attributes are unset, Cygwin falls back to the user's local profile directory, typically something along the lines of C:\Users\$USERNAME. Of course, the Windows directory is converted to POSIX-style by Cygwin. Check with `getent passwd $USER'. Your home dir from the Cygwin POV is what *Windows* sets, not your personal $HOME setting from the environment. I bet you'll see //corp.ad.broadcom.com/DENA/home/reisert in the getent output. By setting $HOME to a directory different from your home dir in the passwd entry, you're confusing your tools. Some of them will happily use $HOME, others will use the entry from the account DB (what getent returns). The bottom line is, don't do that. You have enough configuration methods in /etc/nsswitch.conf, AD, or SAM to set your home dir correctly. Setting $HOME manually to a differnet value is just asking for trouble. HTH, Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat pgpC2lIHF8vPa.pgp Description: PGP signature
Re: Running tasklist /m in cygwin hangs
On Wed, Apr 15, 2015 at 1:52 PM, Saurabh T saur...@hotmail.com wrote: Hi, Running tasklist /m file.dll hangs in Cygwin even though it works perfectly fine in the cmd window. Is there any reason for this? I am using a somewhat older cygwin (1.7.25) on a Windows 7 box, and do not want to upgrade unless newer cygwin doesnt show the problem. Thank you. -- 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 It doesn't hang for me, but it is significantly slower. I'm running the latest development release. Tasklist appears to be WMI heavy. I ran Process Monitor and it appears that when running tasklist /m under bash, WMI is significantly more active. WMI actually opens tzres.dll (Time Zone Resources) about 150,000 times vice a mere 800 times when executing through a standard command prompt. Moreover, WMI really doesn't appear to be doing much of anything with tzres.dll... just open/close/open/close/open/close. Very curious indeed. -- 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: MATE category ?
On Fri, Apr 17, 2015 at 5:57 PM, Marco Atzeri marco.atz...@gmail.com wrote: Yaakov, there are several device with MATE as category. Are we going over the list defined in: https://cygwin.com/setup.html Regards Marco There are also dozens of packages spread out over these categories that don't exist on that list either: KDE LXDE Xfce IMHO, all 4 categories should be made official. Lots of Linux distros use them as categories. The alternative is to put things like window managers and mate-settings-daemon under X11. -Mike
