Not able to full sync Ropsten network past block 6491959 #22497
Description
System information
Geth version: Gemini Sigma (v1.10.1)
OS & Version: Debian GNU/Linux 9.6 (stretch)
Commit hash : c2d2f4e
Actual behaviour
I'm syncing Ropsten network from scratch in full mode. I'm not able to get past block 6491959 (found using web3.eth.syncing). Geth stucks there, sometimes it crashes due to out of memory error (please see the backtrace below) sometimes it just restarts syncing from around block 6488206. And then it gets stuck again at block 6491959, restarting the cycle.
Steps to reproduce the behaviour
I've updated Geth to the latest version and then restarted the node. I've also updated Go to version go1.16 linux/amd64.
Backtrace
Mar 15 09:05:28 hostname geth[24509]: fatal error: runtime: out of memory
Mar 15 09:05:28 hostname geth[24509]: runtime stack:
Mar 15 09:05:28 hostname geth[24509]: runtime.throw(0x133b12c, 0x16)
Mar 15 09:05:28 hostname geth[24509]: runtime/panic.go:1117 +0x72
Mar 15 09:05:28 hostname geth[24509]: runtime.sysMap(0xc3b0000000, 0x40000000, 0x208b770)
Mar 15 09:05:28 hostname geth[24509]: runtime/mem_linux.go:169 +0xc6
Mar 15 09:05:28 hostname geth[24509]: runtime.(*mheap).sysAlloc(0x20714a0, 0x3d000000, 0x443277, 0x20714a8)
Mar 15 09:05:28 hostname geth[24509]: runtime/malloc.go:729 +0x1e5
Mar 15 09:05:28 hostname geth[24509]: runtime.(*mheap).grow(0x20714a0, 0x1e620, 0x0)
Mar 15 09:05:28 hostname geth[24509]: runtime/mheap.go:1346 +0x85
Mar 15 09:05:28 hostname geth[24509]: runtime.(*mheap).allocSpan(0x20714a0, 0x1e620, 0x100, 0x1ff)
Mar 15 09:05:28 hostname geth[24509]: runtime/mheap.go:1173 +0x609
Mar 15 09:05:28 hostname geth[24509]: runtime.(*mheap).alloc.func1()
Mar 15 09:05:28 hostname geth[24509]: runtime/mheap.go:910 +0x59
Mar 15 09:05:28 hostname geth[24509]: runtime.systemstack(0x0)
Mar 15 09:05:28 hostname geth[24509]: runtime/asm_amd64.s:379 +0x66
Mar 15 09:05:28 hostname geth[24509]: runtime.mstart()
Mar 15 09:05:28 hostname geth[24509]: runtime/proc.go:1246