public inbox for ecos-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug 1001633] New: DHCP Client may hang
@ 2012-07-26  9:22 bugzilla-daemon
  0 siblings, 0 replies; 2+ messages in thread
From: bugzilla-daemon @ 2012-07-26  9:22 UTC (permalink / raw)
  To: unassigned

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001633

           Summary: DHCP Client may hang
           Product: eCos
           Version: CVS
          Platform: All
        OS/Version: All
            Status: UNCONFIRMED
          Severity: normal
          Priority: normal
         Component: Ethernet
        AssignedTo: unassigned@bugs.ecos.sourceware.org
        ReportedBy: bernd.edlinger@hotmail.de
                CC: ecos-bugs@ecos.sourceware.org
             Class: Advice Request


Bernd Edlinger <bernd.edlinger@hotmail.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
               Flag|                            |Patch_or_Contribution+

Created an attachment (id=1847)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1847)
proposed patch to fix this defect

when a DHCP Response is received it should contain a time slice value.

however if the time slice value is zero, the device enters an endless loop.
Hence, this configuration error in the DHCP server crashes the devices.


Reproduce:
install tftpd32 on an own subnet, and configure the dhcp server
to use zero-length time slices.
The time slice value is not checked by the dhcp server, and just used as is.
When the eCos DHCP Client is booted, it gets an address assigned, and freezes
immediately, for instance no ping or arp responses are sent.


Additionally, the code that sends new dhcp requests
uses a wrong formula for bp_secs:

            bp_secs = cyg_current_time() / 100;

But that is only correct for 100 Ticks/second.

            bp_secs = cyg_current_time() /
 ((1000000000LL*CYGNUM_HAL_RTC_DENOMINATOR)/CYGNUM_HAL_RTC_NUMERATOR);


Howto Fix:
Attached you will find a patch for the file
"ecos/packages/net/common/current/src/dhcp_prot.c"
which checks for zero timeout and silently changes this to 1 second timeout.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug.


^ permalink raw reply	[flat|nested] 2+ messages in thread

* [Bug 1001633] New: DHCP Client may hang
@ 2012-07-26  9:22 bugzilla-daemon
  0 siblings, 0 replies; 2+ messages in thread
From: bugzilla-daemon @ 2012-07-26  9:22 UTC (permalink / raw)
  To: ecos-bugs

Please do not reply to this email. Use the web interface provided at:
http://bugs.ecos.sourceware.org/show_bug.cgi?id=1001633

           Summary: DHCP Client may hang
           Product: eCos
           Version: CVS
          Platform: All
        OS/Version: All
            Status: UNCONFIRMED
          Severity: normal
          Priority: normal
         Component: Ethernet
        AssignedTo: unassigned@bugs.ecos.sourceware.org
        ReportedBy: bernd.edlinger@hotmail.de
                CC: ecos-bugs@ecos.sourceware.org
             Class: Advice Request


Bernd Edlinger <bernd.edlinger@hotmail.de> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
               Flag|                            |Patch_or_Contribution+

Created an attachment (id=1847)
 --> (http://bugs.ecos.sourceware.org/attachment.cgi?id=1847)
proposed patch to fix this defect

when a DHCP Response is received it should contain a time slice value.

however if the time slice value is zero, the device enters an endless loop.
Hence, this configuration error in the DHCP server crashes the devices.


Reproduce:
install tftpd32 on an own subnet, and configure the dhcp server
to use zero-length time slices.
The time slice value is not checked by the dhcp server, and just used as is.
When the eCos DHCP Client is booted, it gets an address assigned, and freezes
immediately, for instance no ping or arp responses are sent.


Additionally, the code that sends new dhcp requests
uses a wrong formula for bp_secs:

            bp_secs = cyg_current_time() / 100;

But that is only correct for 100 Ticks/second.

            bp_secs = cyg_current_time() /
 ((1000000000LL*CYGNUM_HAL_RTC_DENOMINATOR)/CYGNUM_HAL_RTC_NUMERATOR);


Howto Fix:
Attached you will find a patch for the file
"ecos/packages/net/common/current/src/dhcp_prot.c"
which checks for zero timeout and silently changes this to 1 second timeout.

-- 
Configure bugmail: http://bugs.ecos.sourceware.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.


^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2012-07-26  9:22 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-07-26  9:22 [Bug 1001633] New: DHCP Client may hang bugzilla-daemon
  -- strict thread matches above, loose matches on Subject: below --
2012-07-26  9:22 bugzilla-daemon

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).