bugfix: we were retrying the same circuit after getting a resolve
failure. so of course the next two tries would fail too. now we try
a new circuit each time (at most three times).


svn:r1867

svn:r1866

svn:r1865

re-disable it.


svn:r1864

 * \file circuitbuild.c
 * \brief The actual details of building circuits.

 * \file circuitlist.c
 * \brief Manage the global circuit list.

 * \file circuituse.c
 * \brief Launch the right sort of circuits, attach streams to them.

 * \file connection_edge.c
 * \brief Handle edge streams.

 * \file onion.c
 * \brief Functions to queue create cells, and handle onionskin
 * parsing and creation.

 * \file relay.c
 * \brief Handle relay cell encryption/decryption, plus packaging and
 * receiving from circuits.


svn:r1863

svn:r1862

svn:r1861

svn:r1860

svn:r1859

svn:r1858

svn:r1857

svn:r1856

svn:r1855

svn:r1854

svn:r1853

Make "connected" a different case from "finished_flushing"; always close_immediate whhen connect() fails.


svn:r1852

svn:r1851

svn:r1850

svn:r1849

svn:r1848

svn:r1847

svn:r1846

svn:r1845

svn:r1844

svn:r1843

svn:r1842

svn:r1841

svn:r1840

svn:r1839

svn:r1838

svn:r1837

svn:r1836

svn:r1835

svn:r1834

svn:r1833

svn:r1832

svn:r1831

svn:r1830

svn:r1829

svn:r1828