User Tools

Site Tools


Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
mission:log:2016:03:18:harder-soft-unbricking-a-bricked-ubiquiti-unifi-uap-pro-ap [2016/03/20 07:07] – [Harder Soft-Unbricking a Ubiquiti Unifi UAP-Pro AP] chronomission:log:2016:03:18:harder-soft-unbricking-a-bricked-ubiquiti-unifi-uap-pro-ap [2016/06/09 20:48] (current) – [Solution 1] chrono
Line 58: Line 58:
 </WRAP> </WRAP>
  
-This may be due to the fact that it expects a bricked UBNT firmware and NOT an OpenWRT installation.+This may be due to the fact that this workflow expects a bricked UBNT firmware and NOT an OpenWRT installation.
 ===== Solution 2 ===== ===== Solution 2 =====
  
Line 178: Line 178:
 </code> </code>
  
-The firmware image is stored in RAM now and occupies **0xf60000** bytes. Writing it from there to flash memory directly may fail when the target area has not been erased or if it is write-protected. Let's take care of that first.+The firmware image is stored in RAM now and occupies **0xf60000** bytes. Writing it from there to flash memory directly may fail when the target area has not been erased or is write-protected. Let's take care of that first.
 ==== Step 4: Reset partition table to defaults ==== ==== Step 4: Reset partition table to defaults ====
  
Line 238: Line 238:
 ==== Step 6: Erase the corrupt jffs2 ==== ==== Step 6: Erase the corrupt jffs2 ====
  
-Figuring out the correct address and length was a bit difficult, uncertain, not well documented and took us quite a while, because we didn't want to brick it more by writing stuff somewhere other than we intended. Unfortunately, there was no documentation about the process of how we derived at that conclusion, but in the end we finally decided to start at address **0x9f050000** for the length of the jffs2 image as indicated by the tftp transfer (**0xf60000**) and the mtd partition size (**0xf60000**)+Figuring out the correct address and length was a bit difficult, unclear, uncertain, not well documented and took us quite a while, because we didn't want to brick it more by writing stuff somewhere other than we actually intended. Unfortunately, there was no documentation during the process of how we derived at that conclusion, but in the end we finally decided to start at address **0x9f050000** for the length of the jffs2 image as indicated by the tftp transfer (**0xf60000**) and the mtd partition size (**0xf60000**)
  
 <code> <code>